|
00:30
toggle elapsed time, timecode and framecount 00:30
toggle elapsed time, timecode and framecount
00:30
toggle duration and remaining time
rewind
play/pause
fast forward
show external playback menu
show external playback menu
mute
document actions
soundcore AeroClip (input)_2026-04-18_10-36-56.mp4...
|
QuickTime Player
|
soundcore AeroClip (input)_2026-04-18_10-36-56.mp4
|
NULL
|
|
00:30
toggle elapsed time, timecode and framecount 00:30
toggle elapsed time, timecode and framecount
00:30
toggle duration and remaining time
rewind
play/pause
fast forward
show external playback menu
show external playback menu
mute
document actions
soundcore AeroClip (input)_2026-04-18_10-36-56.mp4...
|
QuickTime Player
|
soundcore AeroClip (input)_2026-04-18_10-36-56.mp4
|
NULL
|
|
00:30
toggle elapsed time, timecode and framecount 00:30
toggle elapsed time, timecode and framecount
00:30
toggle duration and remaining time
rewind
play/pause
fast forward
show external playback menu
show external playback menu
mute
document actions
soundcore AeroClip (input)_2026-04-18_10-36-56.mp4...
|
QuickTime Player
|
soundcore AeroClip (input)_2026-04-18_10-36-56.mp4
|
NULL
|
|
00:00
toggle elapsed time, timecode and framecount 00:00
toggle elapsed time, timecode and framecount
00:30
toggle duration and remaining time
rewind
play/pause
fast forward
show external playback menu
show external playback menu
mute
document actions
soundcore AeroClip (input)_2026-04-18_10-34-24.mp4...
|
QuickTime Player
|
soundcore AeroClip (input)_2026-04-18_10-34-24.mp4
|
NULL
|
|
00:00
toggle elapsed time, timecode and framecount 00:00
toggle elapsed time, timecode and framecount
00:30
toggle duration and remaining time
rewind
play/pause
fast forward
show external playback menu
show external playback menu
mute
document actions
soundcore AeroClip (input)_2026-04-18_10-34-24.mp4...
|
QuickTime Player
|
soundcore AeroClip (input)_2026-04-18_10-34-24.mp4
|
NULL
|
|
Last login: Thu Apr 23 14:01:29 on ttys009
Poetry Last login: Thu Apr 23 14:01:29 on ttys009
Poetry could not find a pyproject.toml file in /Users/lukas or its parents
Poetry could not find a pyproject.toml file in /Users/lukas or its parents
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ cd ~/.screenpipe/data/data
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe/data/data $ cd 2026-04-23
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe/data/data/2026-04-23 $ ll
total 338296
drwxr-xr-x 106 lukas staff 3392 24 Apr 09:20 .
drwxr-xr-x 17 lukas staff 544 24 Apr 09:19 ..
-rw-r--r-- 1 lukas staff 174550 23 Apr 11:38 1776933481776_m1.jpg
-rw-r--r-- 1 lukas staff 1448572 23 Apr 09:24 compact_monitor_1_1776925466477.mp4
-rw-r--r-- 1 lukas staff 2157204 23 Apr 09:29 compact_monitor_1_1776925771961.mp4
-rw-r--r-- 1 lukas staff 785851 23 Apr 09:34 compact_monitor_1_1776926078542.mp4
-rw-r--r-- 1 lukas staff 319924 23 Apr 09:39 compact_monitor_1_1776926389122.mp4
-rw-r--r-- 1 lukas staff 178437 23 Apr 09:44 compact_monitor_1_1776926696343.mp4
-rw-r--r-- 1 lukas staff 182316 23 Apr 09:50 compact_monitor_1_1776927001989.mp4
-rw-r--r-- 1 lukas staff 475405 23 Apr 09:55 compact_monitor_1_1776927311821.mp4
-rw-r--r-- 1 lukas staff 3670216 23 Apr 10:00 compact_monitor_1_1776927616817.mp4
-rw-r--r-- 1 lukas staff 1311845 23 Apr 10:05 compact_monitor_1_1776927930717.mp4
-rw-r--r-- 1 lukas staff 1768745 23 Apr 10:10 compact_monitor_1_1776928242002.mp4
-rw-r--r-- 1 lukas staff 942473 23 Apr 10:15 compact_monitor_1_1776928549085.mp4
-rw-r--r-- 1 lukas staff 911130 23 Apr 10:20 compact_monitor_1_1776928854473.mp4
-rw-r--r-- 1 lukas staff 1249531 23 Apr 10:26 compact_monitor_1_1776929165912.mp4
-rw-r--r-- 1 lukas staff 2106654 23 Apr 10:31 compact_monitor_1_1776929476115.mp4
-rw-r--r-- 1 lukas staff 1633583 23 Apr 10:36 compact_monitor_1_1776929781778.mp4
-rw-r--r-- 1 lukas staff 1186578 23 Apr 10:41 compact_monitor_1_1776930087801.mp4
-rw-r--r-- 1 lukas staff 322993 23 Apr 10:46 compact_monitor_1_1776930392335.mp4
-rw-r--r-- 1 lukas staff 120434 23 Apr 10:51 compact_monitor_1_1776930696294.mp4
-rw-r--r-- 1 lukas staff 111489 23 Apr 10:56 compact_monitor_1_1776931004441.mp4
-rw-r--r-- 1 lukas staff 560588 23 Apr 11:01 compact_monitor_1_1776931314879.mp4
-rw-r--r-- 1 lukas staff 2146373 23 Apr 11:07 compact_monitor_1_1776931618003.mp4
-rw-r--r-- 1 lukas staff 1276021 23 Apr 11:12 compact_monitor_1_1776931930806.mp4
-rw-r--r-- 1 lukas staff 274120 23 Apr 11:17 compact_monitor_1_1776932235717.mp4
-rw-r--r-- 1 lukas staff 212162 23 Apr 11:22 compact_monitor_1_1776932541353.mp4
-rw-r--r-- 1 lukas staff 639071 23 Apr 11:27 compact_monitor_1_1776932847264.mp4
-rw-r--r-- 1 lukas staff 315645 23 Apr 11:32 compact_monitor_1_1776933174068.mp4
-rw-r--r-- 1 lukas staff 518635 23 Apr 11:38 compact_monitor_1_1776933492233.mp4
-rw-r--r-- 1 lukas staff 1886880 23 Apr 11:43 compact_monitor_1_1776933806263.mp4
-rw-r--r-- 1 lukas staff 2316900 23 Apr 11:48 compact_monitor_1_1776934133065.mp4
-rw-r--r-- 1 lukas staff 1589469 23 Apr 11:54 compact_monitor_1_1776934449640.mp4
-rw-r--r-- 1 lukas staff 592352 23 Apr 11:59 compact_monitor_1_1776934755385.mp4
-rw-r--r-- 1 lukas staff 1631938 23 Apr 12:04 compact_monitor_1_1776935058652.mp4
-rw-r--r-- 1 lukas staff 6099603 23 Apr 12:24 compact_monitor_1_1776936289062.mp4
-rw-r--r-- 1 lukas staff 91125 23 Apr 12:35 compact_monitor_1_1776936919782.mp4
-rw-r--r-- 1 lukas staff 2955581 23 Apr 12:40 compact_monitor_1_1776937224602.mp4
-rw-r--r-- 1 lukas staff 3002373 23 Apr 12:45 compact_monitor_1_1776937538337.mp4
-rw-r--r-- 1 lukas staff 3116117 23 Apr 12:50 compact_monitor_1_1776937842939.mp4
-rw-r--r-- 1 lukas staff 2426878 23 Apr 12:55 compact_monitor_1_1776938150380.mp4
-rw-r--r-- 1 lukas staff 2028056 23 Apr 13:01 compact_monitor_1_1776938459676.mp4
-rw-r--r-- 1 lukas staff 2104678 23 Apr 13:06 compact_monitor_1_1776938769006.mp4
-rw-r--r-- 1 lukas staff 3004512 23 Apr 13:11 compact_monitor_1_1776939078771.mp4
-rw-r--r-- 1 lukas staff 1643140 23 Apr 13:16 compact_monitor_1_1776939389326.mp4
-rw-r--r-- 1 lukas staff 1923058 23 Apr 13:21 compact_monitor_1_1776939706398.mp4
-rw-r--r-- 1 lukas staff 900757 23 Apr 13:26 compact_monitor_1_1776940011813.mp4
-rw-r--r-- 1 lukas staff 2964075 23 Apr 13:32 compact_monitor_1_1776940319106.mp4
-rw-r--r-- 1 lukas staff 2419466 23 Apr 13:37 compact_monitor_1_1776940628353.mp4
-rw-r--r-- 1 lukas staff 2447997 23 Apr 13:42 compact_monitor_1_1776940941730.mp4
-rw-r--r-- 1 lukas staff 2202220 23 Apr 13:47 compact_monitor_1_1776941248366.mp4
-rw-r--r-- 1 lukas staff 2309716 23 Apr 13:52 compact_monitor_1_1776941558220.mp4
-rw-r--r-- 1 lukas staff 2049791 23 Apr 13:57 compact_monitor_1_1776941868192.mp4
-rw-r--r-- 1 lukas staff 6097615 24 Apr 09:20 compact_monitor_1_1777011631828.mp4
-rw-r--r-- 1 lukas staff 215938 23 Apr 09:24 compact_monitor_2_1776925467316.mp4
-rw-r--r-- 1 lukas staff 3956084 23 Apr 09:29 compact_monitor_2_1776925774333.mp4
-rw-r--r-- 1 lukas staff 3998164 23 Apr 09:34 compact_monitor_2_1776926081685.mp4
-rw-r--r-- 1 lukas staff 2446471 23 Apr 09:39 compact_monitor_2_1776926391115.mp4
-rw-r--r-- 1 lukas staff 776043 23 Apr 09:44 compact_monitor_2_1776926697217.mp4
-rw-r--r-- 1 lukas staff 802594 23 Apr 09:50 compact_monitor_2_1776927003100.mp4
-rw-r--r-- 1 lukas staff 1453164 23 Apr 09:55 compact_monitor_2_1776927313533.mp4
-rw-r--r-- 1 lukas staff 1449750 23 Apr 10:00 compact_monitor_2_1776927621638.mp4
-rw-r--r-- 1 lukas staff 1635019 23 Apr 10:05 compact_monitor_2_1776927933130.mp4
-rw-r--r-- 1 lukas staff 595843 23 Apr 10:10 compact_monitor_2_1776928245303.mp4
-rw-r--r-- 1 lukas staff 531088 23 Apr 10:15 compact_monitor_2_1776928550520.mp4
-rw-r--r-- 1 lukas staff 846515 23 Apr 10:20 compact_monitor_2_1776928856922.mp4
-rw-r--r-- 1 lukas staff 904088 23 Apr 10:26 compact_monitor_2_1776929169922.mp4
-rw-r--r-- 1 lukas staff 786829 23 Apr 10:31 compact_monitor_2_1776929479675.mp4
-rw-r--r-- 1 lukas staff 729181 23 Apr 10:36 compact_monitor_2_1776929783409.mp4
-rw-r--r-- 1 lukas staff 1116843 23 Apr 10:41 compact_monitor_2_1776930090240.mp4
-rw-r--r-- 1 lukas staff 889822 23 Apr 10:46 compact_monitor_2_1776930393454.mp4
-rw-r--r-- 1 lukas staff 444250 23 Apr 10:51 compact_monitor_2_1776930697139.mp4
-rw-r--r-- 1 lukas staff 1024242 23 Apr 10:56 compact_monitor_2_1776931006436.mp4
-rw-r--r-- 1 lukas staff 1206096 23 Apr 11:01 compact_monitor_2_1776931316182.mp4
-rw-r--r-- 1 lukas staff 1249893 23 Apr 11:07 compact_monitor_2_1776931621999.mp4
-rw-r--r-- 1 lukas staff 401526 23 Apr 11:12 compact_monitor_2_1776931932870.mp4
-rw-r--r-- 1 lukas staff 982256 23 Apr 11:17 compact_monitor_2_1776932236802.mp4
-rw-r--r-- 1 lukas staff 1131582 23 Apr 11:22 compact_monitor_2_1776932542205.mp4
-rw-r--r-- 1 lukas staff 808822 23 Apr 11:27 compact_monitor_2_1776932849317.mp4
-rw-r--r-- 1 lukas staff 442538 23 Apr 11:32 compact_monitor_2_1776933176662.mp4
-rw-r--r-- 1 lukas staff 675274 23 Apr 11:38 compact_monitor_2_1776933495384.mp4
-rw-r--r-- 1 lukas staff 1534516 23 Apr 11:43 compact_monitor_2_1776933812317.mp4
-rw-r--r-- 1 lukas staff 1657590 23 Apr 11:49 compact_monitor_2_1776934138529.mp4
-rw-r--r-- 1 lukas staff 1222887 23 Apr 11:54 compact_monitor_2_1776934453154.mp4
-rw-r--r-- 1 lukas staff 1003999 23 Apr 11:59 compact_monitor_2_1776934756922.mp4
-rw-r--r-- 1 lukas staff 1168918 23 Apr 12:04 compact_monitor_2_1776935062292.mp4
-rw-r--r-- 1 lukas staff 5755378 23 Apr 12:25 compact_monitor_2_1776936297908.mp4
-rw-r--r-- 1 lukas staff 302577 23 Apr 12:25 compact_monitor_2_1776936305094.mp4
-rw-r--r-- 1 lukas staff 1710485 23 Apr 12:35 compact_monitor_2_1776936920837.mp4
-rw-r--r-- 1 lukas staff 2705135 23 Apr 12:40 compact_monitor_2_1776937228721.mp4
-rw-r--r-- 1 lukas staff 1751314 23 Apr 12:45 compact_monitor_2_1776937539934.mp4
-rw-r--r-- 1 lukas staff 924709 23 Apr 12:50 compact_monitor_2_1776937844858.mp4
-rw-r--r-- 1 lukas staff 1285623 23 Apr 12:55 compact_monitor_2_1776938151831.mp4
-rw-r--r-- 1 lukas staff 405857 23 Apr 13:01 compact_monitor_2_1776938461195.mp4
-rw-r--r-- 1 lukas staff 407926 23 Apr 13:06 compact_monitor_2_1776938770290.mp4
-rw-r--r-- 1 lukas staff 958635 23 Apr 13:11 compact_monitor_2_1776939080379.mp4
-rw-r--r-- 1 lukas staff 3622447 23 Apr 13:16 compact_monitor_2_1776939392822.mp4
-rw-r--r-- 1 lukas staff 1850262 23 Apr 13:21 compact_monitor_2_1776939708714.mp4
-rw-r--r-- 1 lukas staff 4725657 23 Apr 13:26 compact_monitor_2_1776940013773.mp4
-rw-r--r-- 1 lukas staff 3590913 23 Apr 13:32 compact_monitor_2_1776940321454.mp4
-rw-r--r-- 1 lukas staff 4073703 23 Apr 13:37 compact_monitor_2_1776940630759.mp4
-rw-r--r-- 1 lukas staff 6432165 23 Apr 13:42 compact_monitor_2_1776940943551.mp4
-rw-r--r-- 1 lukas staff 1484957 23 Apr 13:47 compact_monitor_2_1776941250256.mp4
-rw-r--r-- 1 lukas staff 1336470 23 Apr 13:52 compact_monitor_2_1776941561295.mp4
-rw-r--r-- 1 lukas staff 1327781 23 Apr 13:57 compact_monitor_2_1776941869906.mp4
-rw-r--r-- 1 lukas staff 4219671 24 Apr 09:20 compact_monitor_2_1777011640295.mp4
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe/data/data/2026-04-23 $ cd 2026-04-23
cd: no such file or directory: 2026-04-23
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe/data/data/2026-04-23 $ ll
total 338296
drwxr-xr-x 106 lukas staff 3392 24 Apr 09:20 .
drwxr-xr-x 17 lukas staff 544 24 Apr 09:19 ..
-rw-r--r-- 1 lukas staff 174550 23 Apr 11:38 1776933481776_m1.jpg
-rw-r--r-- 1 lukas staff 1448572 23 Apr 09:24 compact_monitor_1_1776925466477.mp4
-rw-r--r-- 1 lukas staff 2157204 23 Apr 09:29 compact_monitor_1_1776925771961.mp4
-rw-r--r-- 1 lukas staff 785851 23 Apr 09:34 compact_monitor_1_1776926078542.mp4
-rw-r--r-- 1 lukas staff 319924 23 Apr 09:39 compact_monitor_1_1776926389122.mp4
-rw-r--r-- 1 lukas staff 178437 23 Apr 09:44 compact_monitor_1_1776926696343.mp4
-rw-r--r-- 1 lukas staff 182316 23 Apr 09:50 compact_monitor_1_1776927001989.mp4
-rw-r--r-- 1 lukas staff 475405 23 Apr 09:55 compact_monitor_1_1776927311821.mp4
-rw-r--r-- 1 lukas staff 3670216 23 Apr 10:00 compact_monitor_1_1776927616817.mp4
-rw-r--r-- 1 lukas staff 1311845 23 Apr 10:05 compact_monitor_1_1776927930717.mp4
-rw-r--r-- 1 lukas staff 1768745 23 Apr 10:10 compact_monitor_1_1776928242002.mp4
-rw-r--r-- 1 lukas staff 942473 23 Apr 10:15 compact_monitor_1_1776928549085.mp4
-rw-r--r-- 1 lukas staff 911130 23 Apr 10:20 compact_monitor_1_1776928854473.mp4
-rw-r--r-- 1 lukas staff 1249531 23 Apr 10:26 compact_monitor_1_1776929165912.mp4
-rw-r--r-- 1 lukas staff 2106654 23 Apr 10:31 compact_monitor_1_1776929476115.mp4
-rw-r--r-- 1 lukas staff 1633583 23 Apr 10:36 compact_monitor_1_1776929781778.mp4
-rw-r--r-- 1 lukas staff 1186578 23 Apr 10:41 compact_monitor_1_1776930087801.mp4
-rw-r--r-- 1 lukas staff 322993 23 Apr 10:46 compact_monitor_1_1776930392335.mp4
-rw-r--r-- 1 lukas staff 120434 23 Apr 10:51 compact_monitor_1_1776930696294.mp4
-rw-r--r-- 1 lukas staff 111489 23 Apr 10:56 compact_monitor_1_1776931004441.mp4
-rw-r--r-- 1 lukas staff 560588 23 Apr 11:01 compact_monitor_1_1776931314879.mp4
-rw-r--r-- 1 lukas staff 2146373 23 Apr 11:07 compact_monitor_1_1776931618003.mp4
-rw-r--r-- 1 lukas staff 1276021 23 Apr 11:12 compact_monitor_1_1776931930806.mp4
-rw-r--r-- 1 lukas staff 274120 23 Apr 11:17 compact_monitor_1_1776932235717.mp4
-rw-r--r-- 1 lukas staff 212162 23 Apr 11:22 compact_monitor_1_1776932541353.mp4
-rw-r--r-- 1 lukas staff 639071 23 Apr 11:27 compact_monitor_1_1776932847264.mp4
-rw-r--r-- 1 lukas staff 315645 23 Apr 11:32 compact_monitor_1_1776933174068.mp4
-rw-r--r-- 1 lukas staff 518635 23 Apr 11:38 compact_monitor_1_1776933492233.mp4
-rw-r--r-- 1 lukas staff 1886880 23 Apr 11:43 compact_monitor_1_1776933806263.mp4
-rw-r--r-- 1 lukas staff 2316900 23 Apr 11:48 compact_monitor_1_1776934133065.mp4
-rw-r--r-- 1 lukas staff 1589469 23 Apr 11:54 compact_monitor_1_1776934449640.mp4
-rw-r--r-- 1 lukas staff 592352 23 Apr 11:59 compact_monitor_1_1776934755385.mp4
-rw-r--r-- 1 lukas staff 1631938 23 Apr 12:04 compact_monitor_1_1776935058652.mp4
-rw-r--r-- 1 lukas staff 6099603 23 Apr 12:24 compact_monitor_1_1776936289062.mp4
-rw-r--r-- 1 lukas staff 91125 23 Apr 12:35 compact_monitor_1_1776936919782.mp4
-rw-r--r-- 1 lukas staff 2955581 23 Apr 12:40 compact_monitor_1_1776937224602.mp4
-rw-r--r-- 1 lukas staff 3002373 23 Apr 12:45 compact_monitor_1_1776937538337.mp4
-rw-r--r-- 1 lukas staff 3116117 23 Apr 12:50 compact_monitor_1_1776937842939.mp4
-rw-r--r-- 1 lukas staff 2426878 23 Apr 12:55 compact_monitor_1_1776938150380.mp4
-rw-r--r-- 1 lukas staff 2028056 23 Apr 13:01 compact_monitor_1_1776938459676.mp4
-rw-r--r-- 1 lukas staff 2104678 23 Apr 13:06 compact_monitor_1_1776938769006.mp4
-rw-r--r-- 1 lukas staff 3004512 23 Apr 13:11 compact_monitor_1_1776939078771.mp4
-rw-r--r-- 1 lukas staff 1643140 23 Apr 13:16 compact_monitor_1_1776939389326.mp4
-rw-r--r-- 1 lukas staff 1923058 23 Apr 13:21 compact_monitor_1_1776939706398.mp4
-rw-r--r-- 1 lukas staff 900757 23 Apr 13:26 compact_monitor_1_1776940011813.mp4
-rw-r--r-- 1 lukas staff 2964075 23 Apr 13:32 compact_monitor_1_1776940319106.mp4
-rw-r--r-- 1 lukas staff 2419466 23 Apr 13:37 compact_monitor_1_1776940628353.mp4
-rw-r--r-- 1 lukas staff 2447997 23 Apr 13:42 compact_monitor_1_1776940941730.mp4
-rw-r--r-- 1 lukas staff 2202220 23 Apr 13:47 compact_monitor_1_1776941248366.mp4
-rw-r--r-- 1 lukas staff 2309716 23 Apr 13:52 compact_monitor_1_1776941558220.mp4
-rw-r--r-- 1 lukas staff 2049791 23 Apr 13:57 compact_monitor_1_1776941868192.mp4
-rw-r--r-- 1 lukas staff 6097615 24 Apr 09:20 compact_monitor_1_1777011631828.mp4
-rw-r--r-- 1 lukas staff 215938 23 Apr 09:24 compact_monitor_2_1776925467316.mp4
-rw-r--r-- 1 lukas staff 3956084 23 Apr 09:29 compact_monitor_2_1776925774333.mp4
-rw-r--r-- 1 lukas staff 3998164 23 Apr 09:34 compact_monitor_2_1776926081685.mp4
-rw-r--r-- 1 lukas staff 2446471 23 Apr 09:39 compact_monitor_2_1776926391115.mp4
-rw-r--r-- 1 lukas staff 776043 23 Apr 09:44 compact_monitor_2_1776926697217.mp4
-rw-r--r-- 1 lukas staff 802594 23 Apr 09:50 compact_monitor_2_1776927003100.mp4
-rw-r--r-- 1 lukas staff 1453164 23 Apr 09:55 compact_monitor_2_1776927313533.mp4
-rw-r--r-- 1 lukas staff 1449750 23 Apr 10:00 compact_monitor_2_1776927621638.mp4
-rw-r--r-- 1 lukas staff 1635019 23 Apr 10:05 compact_monitor_2_1776927933130.mp4
-rw-r--r-- 1 lukas staff 595843 23 Apr 10:10 compact_monitor_2_1776928245303.mp4
-rw-r--r-- 1 lukas staff 531088 23 Apr 10:15 compact_monitor_2_1776928550520.mp4
-rw-r--r-- 1 lukas staff 846515 23 Apr 10:20 compact_monitor_2_1776928856922.mp4
-rw-r--r-- 1 lukas staff 904088 23 Apr 10:26 compact_monitor_2_1776929169922.mp4
-rw-r--r-- 1 lukas staff 786829 23 Apr 10:31 compact_monitor_2_1776929479675.mp4
-rw-r--r-- 1 lukas staff 729181 23 Apr 10:36 compact_monitor_2_1776929783409.mp4
-rw-r--r-- 1 lukas staff 1116843 23 Apr 10:41 compact_monitor_2_1776930090240.mp4
-rw-r--r-- 1 lukas staff 889822 23 Apr 10:46 compact_monitor_2_1776930393454.mp4
-rw-r--r-- 1 lukas staff 444250 23 Apr 10:51 compact_monitor_2_1776930697139.mp4
-rw-r--r-- 1 lukas staff 1024242 23 Apr 10:56 compact_monitor_2_1776931006436.mp4
-rw-r--r-- 1 lukas staff 1206096 23 Apr 11:01 compact_monitor_2_1776931316182.mp4
-rw-r--r-- 1 lukas staff 1249893 23 Apr 11:07 compact_monitor_2_1776931621999.mp4
-rw-r--r-- 1 lukas staff 401526 23 Apr 11:12 compact_monitor_2_1776931932870.mp4
-rw-r--r-- 1 lukas staff 982256 23 Apr 11:17 compact_monitor_2_1776932236802.mp4
-rw-r--r-- 1 lukas staff 1131582 23 Apr 11:22 compact_monitor_2_1776932542205.mp4
-rw-r--r-- 1 lukas staff 808822 23 Apr 11:27 compact_monitor_2_1776932849317.mp4
-rw-r--r-- 1 lukas staff 442538 23 Apr 11:32 compact_monitor_2_1776933176662.mp4
-rw-r--r-- 1 lukas staff 675274 23 Apr 11:38 compact_monitor_2_1776933495384.mp4
-rw-r--r-- 1 lukas staff 1534516 23 Apr 11:43 compact_monitor_2_1776933812317.mp4
-rw-r--r-- 1 lukas staff 1657590 23 Apr 11:49 compact_monitor_2_1776934138529.mp4
-rw-r--r-- 1 lukas staff 1222887 23 Apr 11:54 compact_monitor_2_1776934453154.mp4
-rw-r--r-- 1 lukas staff 1003999 23 Apr 11:59 compact_monitor_2_1776934756922.mp4
-rw-r--r-- 1 lukas staff 1168918 23 Apr 12:04 compact_monitor_2_1776935062292.mp4
-rw-r--r-- 1 lukas staff 5755378 23 Apr 12:25 compact_monitor_2_1776936297908.mp4
-rw-r--r-- 1 lukas staff 302577 23 Apr 12:25 compact_monitor_2_1776936305094.mp4
-rw-r--r-- 1 lukas staff 1710485 23 Apr 12:35 compact_monitor_2_1776936920837.mp4
-rw-r--r-- 1 lukas staff 2705135 23 Apr 12:40 compact_monitor_2_1776937228721.mp4
-rw-r--r-- 1 lukas staff 1751314 23 Apr 12:45 compact_monitor_2_1776937539934.mp4
-rw-r--r-- 1 lukas staff 924709 23 Apr 12:50 compact_monitor_2_1776937844858.mp4
-rw-r--r-- 1 lukas staff 1285623 23 Apr 12:55 compact_monitor_2_1776938151831.mp4
-rw-r--r-- 1 lukas staff 405857 23 Apr 13:01 compact_monitor_2_1776938461195.mp4
-rw-r--r-- 1 lukas staff 407926 23 Apr 13:06 compact_monitor_2_1776938770290.mp4
-rw-r--r-- 1 lukas staff 958635 23 Apr 13:11 compact_monitor_2_1776939080379.mp4
-rw-r--r-- 1 lukas staff 3622447 23 Apr 13:16 compact_monitor_2_1776939392822.mp4
-rw-r--r-- 1 lukas staff 1850262 23 Apr 13:21 compact_monitor_2_1776939708714.mp4
-rw-r--r-- 1 lukas staff 4725657 23 Apr 13:26 compact_monitor_2_1776940013773.mp4
-rw-r--r-- 1 lukas staff 3590913 23 Apr 13:32 compact_monitor_2_1776940321454.mp4
-rw-r--r-- 1 lukas staff 4073703 23 Apr 13:37 compact_monitor_2_1776940630759.mp4
-rw-r--r-- 1 lukas staff 6432165 23 Apr 13:42 compact_monitor_2_1776940943551.mp4
-rw-r--r-- 1 lukas staff 1484957 23 Apr 13:47 compact_monitor_2_1776941250256.mp4
-rw-r--r-- 1 lukas staff 1336470 23 Apr 13:52 compact_monitor_2_1776941561295.mp4
-rw-r--r-- 1 lukas staff 1327781 23 Apr 13:57 compact_monitor_2_1776941869906.mp4
-rw-r--r-- 1 lukas staff 4219671 24 Apr 09:20 compact_monitor_2_1777011640295.mp4
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe/data/data/2026-04-23 $ ~/.screenpipe/screenpipe_sync.sh 2026-04-23
[2026-04-24 09:31:53] ========================================
[2026-04-24 09:31:53] Screenpipe sync starting for: 2026-04-23
[2026-04-24 09:31:53] ========================================
[+00m00s] ▶ Preflight checks
Source DB: OK (6.8G)
[2026-04-24 09:31:53] ERROR: NAS not mounted at /Volumes/Test/screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe/data/data/2026-04-23 $ ~/.screenpipe/screenpipe_sync.sh 2026-04-23
[2026-04-24 09:32:09] ========================================
[2026-04-24 09:32:09] Screenpipe sync starting for: 2026-04-23
[2026-04-24 09:32:09] ========================================
[+00m00s] ▶ Preflight checks
Source DB: OK (6.8G)
NAS mount: OK /Volumes/Test/screenpipe
Archive DB: exists (5.8G)
Data dir: OK (104 files, 165M)
[+00m01s] ▶ Counting source rows for 2026-04-23
frames: 2564
elements: 208937
ui_events: 3546
ocr_text: 904
meetings: 2
[+00m01s] ▶ Initialising tables, indexes, FTS
creating tables ✓ 0m01s
creating indexes ⠋
DOCKER
Close Tab
DEV (-zsh)
Close Tab
APP (-zsh)
Close Tab
screenpipe"
Close Tab
sleep
Close Tab
⌥⌘1
sleep...
|
iTerm2
|
sleep
|
NULL
|
|
Last login: Tue Apr 21 09:37:45 on ttys013
DOCKER
Last login: Tue Apr 21 09:37:45 on ttys013
DOCKER
Close Tab
-zsh
Close Tab
-zsh
Close Tab
✳ Build full day activity summary from Screenpipe (node)
Close Tab
screenpipe"
Close Tab
-zsh
Close Tab
APP (-zsh)
Close Tab
ec2-user@ip-10-30-159-186:~ (nc)
Close Tab
sh
Close Tab
⌥⌘1
sh
lermzsnellVIewsessionScriptsProtllesWindowmelpJ JIMINNY83|* Build full day…. 84"Dally - Platrorm • now100% C48• Tue 21 Apr 9:53:01APP (-zsh)ec2-user@ip-10-3... 2889• Search|Last login: Tue Apr 21 09:37:45 on ttys013IY [SRD-6793] Les Mills activity t X[SRD-6787] Issue with reconnectilX Jiminny MCP Connector - Product(JY-20676] Notify the user if a ParM Jiminny Mail(JY-20500) Batch initial sync for SIa Feed — jiminny — Sentry& Jiminny() JY-20701 | Reschedule HubSpot SPipelines - jiminny/appNew Tab— New Tabg For you(• Recenti* Starred0+ Apps|Q Spaces+ ***1 Service-Desk•, Queues+ ...Team Priorityj All open tickets 12•, Unassigned t... 25 Support tea...j Raised by me•, Assianed to ...E Service requ..4Ej Platform team•, Processing t..Site reliability... 0j New features... 0•, InfoSec issues 0bi Readv for Cu... 0j Resolved ti... 999+= View all queues >• Service reauestsA Incidentsll Reports@: Operations• Knowledge Base° Customers• Channels• Email loasI< Developer escalationsSlack integration& Reporting Center[* Add shortcut, backSKU"O/ySCustomer typeEnterpriseActual outcomelNo activity types are being importedExpected outcomeThe activity types to get importedSeverity levelS2ImpactRoot causeVia have no access to FieldDefinition (needed to sync activity types).Linked work itemsrelates to# JY-20698 Les Mills activity tvpes not pulling inActivityCommentsHistory Work log ApprovalsAdd internal note / Reply to customerPro tio: oress M to comment|Lukas Kovalik 1 minute ago Edited & Internal notePulsar seems to be working fine. I created new playbooks, and the fields were populated, includimissing.For e Mile atir yet moy he eae une we e ter syne has ne er me sha no aoceydoesn't seem to have access to the Organization. Therefore, we can't sync team metadata succe1 [2026-04-21 06:39:44] production.ERROR: [Salesforce] Request exception [4et • Edit • DeleteStovan Tomov 5 davs ado A Internal noteFYI. Pulsar Group (EU) is facina the same issue.€ • Edit • Delete...
|
iTerm2
|
sh
|
NULL
|
|
Platform Sprint 2 Q2 - Platform Team - Scrum Board Platform Sprint 2 Q2 - Platform Team - Scrum Board - Jira
[SRD-6793] Les Mills activity types not pulling in - Jira
[SRD-6793] Les Mills activity types not pulling in - Jira
JY-20698 handle failed field sync on playbook import activity types by LakyLak · Pull Request #11988 · jiminny/app
JY-20698 handle failed field sync on playbook import activity types by LakyLak · Pull Request #11988 · jiminny/app
JY-20692 change confirmation parameter by LakyLak · Pull Request #11986 · jiminny/app
JY-20692 change confirmation parameter by LakyLak · Pull Request #11986 · jiminny/app
[JY-20543] AJ Reports > Tracking - Jira
[JY-20543] AJ Reports > Tracking - Jira
[JY-18909] [Part2] Automated reports with Ask Jiminny - Jira
[JY-18909] [Part2] Automated reports with Ask Jiminny - Jira
Ask Jiminny Reports by nikolay-yankov · Pull Request #11894 · jiminny/app
Ask Jiminny Reports by nikolay-yankov · Pull Request #11894 · jiminny/app
New Tab
New Tab
Product Growth Platform | Userpilot
Product Growth Platform | Userpilot
Userpilot | Events
Userpilot | Events
fix(security): composer dependency updates – 2026-04-15 by github-actions[bot] · Pull Request #11970 · jiminny/app
fix(security): composer dependency updates – 2026-04-15 by github-actions[bot] · Pull Request #11970 · jiminny/app
Pipelines - jiminny/app
Pipelines - jiminny/app
sentry - Google Search
sentry - Google Search
Close tab
New Tab
Customize sidebar
Open Google Gemini (⌃X)
Tabs from other devices
Open history (⇧⌘H)
Close bookmarks (⌘B)
Bookmarks
Bookmarks
Close sidebar
Search bookmarks
Skip to main content
Skip to main content
Accessibility help
Accessibility help
Accessibility feedback
Accessibility feedback
Go to Google Home
sentry
sentry
Clear
Search by voice
Search by image
Search
Google apps
Google Account: [EMAIL]
AI Mode
AI Mode
All
All
Images
Images
Videos
Videos
News
News
Forums
Forums
Short videos
Short videos
More filters
More
Tools
Tools
Search Results
Search Results
Sponsored result
Sponsored result
Resolve Application Errors - Unite Metrics, Traces, & Logs Datadog https://www.datadoghq.com › error-tracking › alternative
Resolve Application Errors - Unite Metrics, Traces, & Logs
Resolve Application Errors - Unite Metrics, Traces, & Logs
Datadog
https://www.datadoghq.com › error-tracking › alternative
Why this ad?
Visualize The Distribution Of Errors Across All Tags To Understand The Source Of The Issue. Track...
Datadog Free Trial Sign Up Today For A Free Trial And See Value Immediately.
Datadog Free Trial
Sign Up Today For A Free Trial And See Value Immediately.
Dynamic Monitoring Multidimensional Monitoring Built For Dynamic & Hybrid Architectures.
Dynamic Monitoring
Multidimensional Monitoring Built For Dynamic & Hybrid Architectures.
Logging Without Limits Manage All Logs Cost-Effectively & Only Pay To Index The Logs You Need
Logging Without Limits
Manage All Logs Cost-Effectively & Only Pay To Index The Logs You Need
1,000+ Integrations Datadog Offers And Supports Wide Coverage Across Any Technology.
1,000+ Integrations
Datadog Offers And Supports Wide Coverage Across Any Technology.
Unified Observability See Metrics, Traces, Logs, And More In A Single Pane View.
Unified Observability
See Metrics, Traces, Logs, And More In A Single Pane View.
Hide sponsored result
Hide sponsored result
Sentry
Sentry
Sentry
Comic book character
More options for Sentry
Image of Sentry | Marvel Cinematic Universe Wiki | Fandom
Image of Sentry (Robert Reynolds) - Wikipedia
Image of How do you all feel about the Sentry as a character? Is he ...
The Sentry and The Void are respectively a superhero and supervillain appearing in American comic books published by Marvel Comics. Created by Paul Jenkins and Jae Lee, with uncredited conceptual contributions by Rick Veitch, the characters first appeared in The Sentry #1.
Source:
Wikipedia
Wikipedia
Movie More info about this Thunderbolts*2025
Movie
Thunderbolts*
2025
Played by More info about this Lewis PullmanThunderbolts*
Played by
Lewis Pullman
Thunderbolts*
Web results
Web results
Sentry: Application Performance Monitoring & Error Tracking ... Sentry https://sentry.io › ...
Sentry: Application Performance Monitoring & Error Tracking ...
Sentry: Application Performance Monitoring & Error Tracking ...
Sentry
https://sentry.io
› ...
About this result
Application performance monitoring
for developers & software teams to see errors clearer, solve issues faster & continue learning continuously.
Dictionary
Dictionary
Definitions from
Oxford Languages
Oxford Languages
Learn more
Learn more
Listen
sentry
/
ˈsɛntri
/
noun
noun
:
sentry
;
plural noun
:
sentries
a soldier
stationed
stationed
to keep guard or to control access to a place.
"sentries patrolled the border"
Similar:
guard
guard
sentinel
sentinel
lookout
lookout
watch
watch
watchman
watchman
patrol
patrol
picket
picket
vedette
vedette
Phrases
Phrases
stand sentry
stand sentry
—
keep guard or control access to a place.
Origin
Origin
early 17th century: perhaps from obsolete
centrinel
, variant of
sentinel
sentinel
.
Use over time for: sentry
Use over time for: sentry
Feedback
Feedback
See more
See more
Creators
Paul Jenkins
Paul Jenkins
,
Jae Lee
Jae Lee
,
Rick Veitch
Rick Veitch
Fictional universe
Marvel Universe
Marvel Universe
Notable aliases
Bob, Golden Guardian of Good, Golden Man
·
See more
See more
Powers and abilities
Flight
Flight
,
Superhuman strength
Superhuman strength
,
Immortality
Immortality
·
See more
See more
Species
Human
Human
People also ask
People also ask
Is the Sentry good or evil?
Is the Sentry good or evil?
What does "sentry" mean?
What does "sentry" mean?
What exactly is Sentry in Marvel?
What exactly is Sentry in Marvel?
What does a Sentry do?
What does a Sentry do?
Web results
Web results
Sentry (Robert Reynolds) Wikipedia https://en.wikipedia.org › wiki › Sentry_(Robert_Reyno...
Sentry (Robert Reynolds)
Sentry (Robert Reynolds)
Wikipedia
https://en.wikipedia.org
› wiki › Sentry_(Robert_Reyno...
About this result
The Sentry (Robert "Bob" Reynolds) and The Void are respectively
a superhero and supervillain
appearing in American comic books published by Marvel Comics.
SENTRY Definition & Meaning Merriam-Webster https://www.merriam-webster.com › dictionary › sentry
SENTRY Definition & Meaning
SENTRY Definition & Meaning
Merriam-Webster
https://www.merriam-webster.com
› dictionary › sentry
About this result
Apr 6, 2026
—
The meaning of SENTRY is
guard, watch; especially
: a soldier standing guard at a point of passage (such as a gate). How to use sentry in a ...
Sentry | Marvel Cinematic Universe Wiki - Fandom Marvel Cinematic Universe Wiki https://marvelcinematicuniverse.fandom.com › wiki › Se...
Sentry | Marvel Cinematic Universe Wiki - Fandom
Sentry | Marvel Cinematic Universe Wiki - Fandom
Marvel Cinematic Universe Wiki
https://marvelcinematicuniverse.fandom.com
› wiki › Se...
About this result
Robert "Bob" Reynolds
is a former meth addict who took part in Project Sentry, an operation by O.X.E. designed to create the ultimate superhero.
Sentry Docs | Application Performance Monitoring & Error ... Sentry Docs https://docs.sentry.io
Sentry Docs | Application Performance Monitoring & Error ...
Sentry Docs | Application Performance Monitoring & Error ...
Sentry Docs
https://docs.sentry.io
About this result
Sentry
provides end-to-end distributed tracing, enabling developers to identify and debug performance issues and errors across their systems and services.
Sentry (Robert Reynolds) Powers, Enemies, History Marvel.com https://www.marvel.com › characters › sentry-robert-re...
Sentry (Robert Reynolds) Powers, Enemies, History
Sentry (Robert Reynolds) Powers, Enemies, History
Marvel.com
https://www.marvel.com
› characters › sentry-robert-re...
About this result
The Sentry possesses remarkable levels of
superhuman strength, speed, and intelligence
, as well as invulnerability, enhanced senses, energy-projection ...
getsentry/sentry: Developer-first error tracking and ... GitHub https://github.com › getsentry › sentry
getsentry/sentry: Developer-first error tracking and ...
getsentry/sentry: Developer-first error tracking and ...
GitHub
https://github.com
› getsentry › sentry
About this result
Sentry
is the debugging platform that helps every developer detect, trace, and fix issues. Code breaks, fix it faster.
The Sentry: Home thesentry.org https://thesentry.org
The Sentry: Home
The Sentry: Home
thesentry.org
https://thesentry.org
About this result
The Sentry is
an investigative and policy organization
that seeks to disable multinational predatory networks that benefit from violent conflict, repression, ...
Sign up for Sentry to start your free trial. Sentry https://sentry.io › signup
Sign up for Sentry to start your free trial.
Sign up for Sentry to start your free trial.
Sentry
https://sentry.io
› signup
About this result
Debug any software issue
, onboard your team, and integrate with your systems. You get 14 days free on our Business plan to start — no credit card required.
Who is Sentry? : r/superheroes Reddit · r/superheroes 30+ comments · 11 months ago
Who is Sentry? : r/superheroes
Who is Sentry? : r/superheroes
Reddit · r/superheroes
30+ comments · 11 months ago
About this result
Sentry is a deeply traumatized man with severe psychological issues
. That's why he has an alternate personality called the Void, an evil force that takes ...
Sponsored result
Sponsored result
Resolve Application Errors | Visualize Key Metrics Datadog https://www.datadoghq.com › error-tracking › alternative
Resolve Application Errors | Visualize Key Metrics
Resolve Application Errors | Visualize Key Metrics
Datadog
https://www.datadoghq.com › error-tracking › alternative
Why this ad?
Visualize The Distribution Of Errors Across All Tags To Understand The Source Of The Issue. Track...
Datadog Free Trial
Datadog Free Trial...
|
Firefox
|
sentry - Google Search — Work
|
NULL
|
|
Platform Sprint 2 Q2 - Platform Team - Scrum Board Platform Sprint 2 Q2 - Platform Team - Scrum Board - Jira
[SRD-6793] Les Mills activity types not pulling in - Jira
[SRD-6793] Les Mills activity types not pulling in - Jira
JY-20698 handle failed field sync on playbook import activity types by LakyLak · Pull Request #11988 · jiminny/app
JY-20698 handle failed field sync on playbook import activity types by LakyLak · Pull Request #11988 · jiminny/app
JY-20692 change confirmation parameter by LakyLak · Pull Request #11986 · jiminny/app
JY-20692 change confirmation parameter by LakyLak · Pull Request #11986 · jiminny/app
[JY-20543] AJ Reports > Tracking - Jira
[JY-20543] AJ Reports > Tracking - Jira
[JY-18909] [Part2] Automated reports with Ask Jiminny - Jira
[JY-18909] [Part2] Automated reports with Ask Jiminny - Jira
Ask Jiminny Reports by nikolay-yankov · Pull Request #11894 · jiminny/app
Ask Jiminny Reports by nikolay-yankov · Pull Request #11894 · jiminny/app
New Tab
New Tab
Product Growth Platform | Userpilot
Product Growth Platform | Userpilot
Userpilot | Events
Userpilot | Events
fix(security): composer dependency updates – 2026-04-15 by github-actions[bot] · Pull Request #11970 · jiminny/app
fix(security): composer dependency updates – 2026-04-15 by github-actions[bot] · Pull Request #11970 · jiminny/app
Pipelines - jiminny/app
Pipelines - jiminny/app
sentry - Google Search
sentry - Google Search
Close tab
New Tab
Customize sidebar
Open Google Gemini (⌃X)
Tabs from other devices
Open history (⇧⌘H)
Close bookmarks (⌘B)
Bookmarks
Bookmarks
Close sidebar
Search bookmarks
Skip to main content
Skip to main content
Accessibility help
Accessibility help
Accessibility feedback
Accessibility feedback
Go to Google Home
sentry
sentry
Clear
Search by voice
Search by image
Search
Google apps
Google Account: [EMAIL]
AI Mode
AI Mode
All
All
Images
Images
Videos
Videos
News
News
Forums
Forums
Short videos
Short videos
More filters
More
Tools
Tools
Search Results
Search Results
Sponsored result
Sponsored result
Resolve Application Errors - Unite Metrics, Traces, & Logs Datadog https://www.datadoghq.com › error-tracking › alternative
Resolve Application Errors - Unite Metrics, Traces, & Logs
Resolve Application Errors - Unite Metrics, Traces, & Logs
Datadog
https://www.datadoghq.com › error-tracking › alternative
Why this ad?
Visualize The Distribution Of Errors Across All Tags To Understand The Source Of The Issue. Track...
Datadog Free Trial Sign Up Today For A Free Trial And See Value Immediately.
Datadog Free Trial
Sign Up Today For A Free Trial And See Value Immediately.
Dynamic Monitoring Multidimensional Monitoring Built For Dynamic & Hybrid Architectures.
Dynamic Monitoring
Multidimensional Monitoring Built For Dynamic & Hybrid Architectures.
Logging Without Limits Manage All Logs Cost-Effectively & Only Pay To Index The Logs You Need
Logging Without Limits
Manage All Logs Cost-Effectively & Only Pay To Index The Logs You Need
1,000+ Integrations Datadog Offers And Supports Wide Coverage Across Any Technology.
1,000+ Integrations
Datadog Offers And Supports Wide Coverage Across Any Technology.
Unified Observability See Metrics, Traces, Logs, And More In A Single Pane View.
Unified Observability
See Metrics, Traces, Logs, And More In A Single Pane View.
Hide sponsored result
Hide sponsored result
Sentry
Sentry
Sentry
Comic book character
More options for Sentry
Image of Sentry | Marvel Cinematic Universe Wiki | Fandom
Image of Sentry (Robert Reynolds) - Wikipedia
Image of How do you all feel about the Sentry as a character? Is he ...
The Sentry and The Void are respectively a superhero and supervillain appearing in American comic books published by Marvel Comics. Created by Paul Jenkins and Jae Lee, with uncredited conceptual contributions by Rick Veitch, the characters first appeared in The Sentry #1.
Source:
Wikipedia
Wikipedia
Movie More info about this Thunderbolts*2025
Movie
Thunderbolts*
2025
Played by More info about this Lewis PullmanThunderbolts*
Played by
Lewis Pullman
Thunderbolts*
Web results
Web results
Sentry: Application Performance Monitoring & Error Tracking ... Sentry https://sentry.io › ...
Sentry: Application Performance Monitoring & Error Tracking ...
Sentry: Application Performance Monitoring & Error Tracking ...
Sentry
https://sentry.io...
|
Firefox
|
sentry - Google Search — Work
|
sentry.io/welcome/
|
|
Platform Sprint 2 Q2 - Platform Team - Scrum Board Platform Sprint 2 Q2 - Platform Team - Scrum Board - Jira
[SRD-6793] Les Mills activity types not pulling in - Jira
[SRD-6793] Les Mills activity types not pulling in - Jira
JY-20698 handle failed field sync on playbook import activity types by LakyLak · Pull Request #11988 · jiminny/app
JY-20698 handle failed field sync on playbook import activity types by LakyLak · Pull Request #11988 · jiminny/app
JY-20692 change confirmation parameter by LakyLak · Pull Request #11986 · jiminny/app
JY-20692 change confirmation parameter by LakyLak · Pull Request #11986 · jiminny/app
[JY-20543] AJ Reports > Tracking - Jira
[JY-20543] AJ Reports > Tracking - Jira
[JY-18909] [Part2] Automated reports with Ask Jiminny - Jira
[JY-18909] [Part2] Automated reports with Ask Jiminny - Jira
Ask Jiminny Reports by nikolay-yankov · Pull Request #11894 · jiminny/app
Ask Jiminny Reports by nikolay-yankov · Pull Request #11894 · jiminny/app
New Tab
New Tab
Product Growth Platform | Userpilot
Product Growth Platform | Userpilot
Userpilot | Events
Userpilot | Events
fix(security): composer dependency updates – 2026-04-15 by github-actions[bot] · Pull Request #11970 · jiminny/app
fix(security): composer dependency updates – 2026-04-15 by github-actions[bot] · Pull Request #11970 · jiminny/app
Pipelines - jiminny/app
Pipelines - jiminny/app
sentry - Google Search
sentry - Google Search
Close tab
New Tab
Customize sidebar
Open Google Gemini (⌃X)
Tabs from other devices
Open history (⇧⌘H)
Close bookmarks (⌘B)
Bookmarks
Bookmarks
Close sidebar
Search bookmarks
Skip to main content
Skip to main content
Accessibility help
Accessibility help
Accessibility feedback
Accessibility feedback
Go to Google Home
sentry
sentry
Clear
Search by voice
Search by image
Search
Google apps
Google Account: [EMAIL]
AI Mode
AI Mode
All
All
Images
Images
Videos
Videos
News
News
Forums
Forums
Short videos
Short videos
More filters
More
Tools
Tools
Search Results
Search Results
Sponsored result
Sponsored result
Resolve Application Errors - Unite Metrics, Traces, & Logs Datadog https://www.datadoghq.com › error-tracking › alternative
Resolve Application Errors - Unite Metrics, Traces, & Logs
Resolve Application Errors - Unite Metrics, Traces, & Logs
Datadog
https://www.datadoghq.com › error-tracking › alternative
Why this ad?
Visualize The Distribution Of Errors Across All Tags To Understand The Source Of The Issue. Track...
Datadog Free Trial Sign Up Today For A Free Trial And See Value Immediately.
Datadog Free Trial
Sign Up Today For A Free Trial And See Value Immediately.
Dynamic Monitoring Multidimensional Monitoring Built For Dynamic & Hybrid Architectures.
Dynamic Monitoring
Multidimensional Monitoring Built For Dynamic & Hybrid Architectures.
Logging Without Limits Manage All Logs Cost-Effectively & Only Pay To Index The Logs You Need
Logging Without Limits
Manage All Logs Cost-Effectively & Only Pay To Index The Logs You Need
1,000+ Integrations Datadog Offers And Supports Wide Coverage Across Any Technology.
1,000+ Integrations
Datadog Offers And Supports Wide Coverage Across Any Technology.
Unified Observability See Metrics, Traces, Logs, And More In A Single Pane View.
Unified Observability
See Metrics, Traces, Logs, And More In A Single Pane View.
Hide sponsored result
Hide sponsored result
Sentry
Sentry
Sentry
Comic book character
More options for Sentry
Image of Sentry | Marvel Cinematic Universe Wiki | Fandom
Image of Sentry (Robert Reynolds) - Wikipedia
Image of How do you all feel about the Sentry as a character? Is he ...
The Sentry and The Void are respectively a superhero and supervillain appearing in American comic books published by Marvel Comics. Created by Paul Jenkins and Jae Lee, with uncredited conceptual contributions by Rick Veitch, the characters first appeared in The Sentry #1.
Source:
Wikipedia
Wikipedia
Movie More info about this Thunderbolts*2025
Movie
Thunderbolts*
2025
Played by More info about this Lewis PullmanThunderbolts*
Played by
Lewis Pullman
Thunderbolts*
Web results
Web results
Sentry: Application Performance Monitoring & Error Tracking ... Sentry https://sentry.io › ...
Sentry: Application Performance Monitoring & Error Tracking ...
Sentry: Application Performance Monitoring & Error Tracking ...
Sentry
https://sentry.io
› ...
About this result
Application performance monitoring
for developers & software teams to see errors clearer, solve issues faster & continue learning continuously.
Dictionary
Dictionary
Definitions from
Oxford Languages
Oxford Languages
Learn more
Learn more
Listen
sentry
/
ˈsɛntri
/
noun
a soldier
stationed
stationed
to keep guard or to control access to a place.
"sentries patrolled the border"
Similar:
guard
guard
sentinel
sentinel
lookout
lookout
watch
watch
watchman
watchman
patrol
patrol
picket
picket
Show All Nyms
Feedback
Feedback
See more
See more
Creators
Paul Jenkins
Paul Jenkins
,
Jae Lee
Jae Lee
,
Rick Veitch
Rick Veitch
Fictional universe
Marvel Universe
Marvel Universe
Notable aliases
Bob, Golden Guardian of Good, Golden Man
·
See more
See more
Powers and abilities
Flight
Flight
,
Superhuman strength
Superhuman strength
,
Immortality
Immortality
·
See more
See more
Species
Human
Human
People also ask
People also ask...
|
Firefox
|
sentry - Google Search — Work
|
sentry.io/welcome/
|
|
Explorer (⇧⌘E)
Search (⇧⌘F)
Source Control (⌃⇧ Explorer (⇧⌘E)
Search (⇧⌘F)
Source Control (⌃⇧G) - 2 pending changes
2
Run and Debug (⇧⌘D)
Remote Explorer
Extensions (⇧⌘X) - 2 require update, 1 requires restart
3
Claude Code
Containers
EXPLORER
EXPLORER
Explorer Section: screenpipe [SSH: nas]
Explorer Section: screenpipe [SSH: nas]
SCREENPIPE [SSH: NAS]
#recycle
app
data
pipes
.gitignore
app_settings.json
archive.db
db.sqlite
screenpipe_sync.sh
screenpipe.db
Outline Section
OUTLINE
OUTLINE
Timeline Section
TIMELINE
TIMELINE
app_settings.json, Editor Group 1
screenpipe_sync.sh, Editor Group 1
db.sqlite-shm, preview, Editor Group 1
#!/bin/bash
# screenpipe_sync.sh
# Syncs Screenpipe SQLite data to a NAS archive database (append-only, no deletions).
# Also copies the day's video/frame data folder to the NAS.
#
# Usage:
# ./screenpipe_sync.sh # syncs yesterday (default)
# ./screenpipe_sync.sh 2026-04-15 # syncs a specific date
# ./screenpipe_sync.sh today # syncs today so far
#
# Cron example (runs at 3am daily):
# 0 3 * * * /Users/lukas/.screenpipe/screenpipe_sync.sh >> /Users/lukas/.screenpipe/sync.log 2>&1
set -euo pipefail
# ─── CONFIG ───────────────────────────────────────────────────────────────────
DB_SRC="${SCREENPIPE_DB:-$HOME/.screenpipe/db.sqlite}"
NAS_MOUNT="${NAS_MOUNT:-/Volumes/Test/screenpipe}"
NAS_DB="$NAS_MOUNT/archive.db"
NAS_DATA="$NAS_MOUNT/data"
LOG_FILE="$HOME/.screenpipe/sync.log"
# ──────────────────────────────────────────────────────────────────────────────
# ─── HELPERS ──────────────────────────────────────────────────────────────────
SCRIPT_START=$(date +%s)
log() {
local msg="[$(date '+%Y-%m-%d %H:%M:%S')] $*"
echo "$msg" | tee -a "$LOG_FILE"
}
step() {
local now=$(date +%s)
local elapsed=$(( now - SCRIPT_START ))
local min=$(( elapsed / 60 ))
local sec=$(( elapsed % 60 ))
printf "\n[+%02dm%02ds] ▶ %s\n" "$min" "$sec" "$*" | tee -a "$LOG_FILE"
}
run_sqlite_heredoc() {
local label="$1"
local sql="$2"
local start=$(date +%s)
printf " %-36s " "$label"
sqlite3 "$DB_SRC" <<< "$sql" &
local pid=$!
local spin=[PASSWORD] '⠙' '⠹' '⠸' '⠼' '⠴' '⠦' '⠧' '⠇' '⠏')
local i=0
while kill -0 "$pid" 2>/dev/null; do
printf "\r %-36s %s " "$label" "${spin[$i]}"
i=$(( (i + 1) % 10 ))
sleep 0.2
done
wait "$pid"
local rc=$?
if [ $rc -ne 0 ]; then
printf "\r %-36s ✗ FAILED\n" "$label" | tee -a "$LOG_FILE"
exit $rc
fi
local dur=$(( $(date +%s) - start ))
printf "\r %-36s ✓ %dm%02ds\n" "$label" "$(( dur / 60 ))" "$(( dur % 60 ))" | tee -a "$LOG_FILE"
}
check() {
local label="$1" got="$2" expected="$3"
if [ "$got" -eq "$expected" ]; then
printf " %-20s %s / %s ✓\n" "$label:" "$got" "$expected"
else
printf " %-20s %s / %s ✗ MISMATCH\n" "$label:" "$got" "$expected"
fi
}
# ──────────────────────────────────────────────────────────────────────────────
# ─── DATE ARGUMENT ────────────────────────────────────────────────────────────
if [ "${1:-}" = "today" ]; then
TARGET_DATE=$(date +%Y-%m-%d)
elif [ -n "${1:-}" ]; then
TARGET_DATE="$1"
if ! [[ "$TARGET_DATE" =~ ^[0-9]{4}-[0-9]{2}-[0-9]{2}$ ]]; then
echo "ERROR: Invalid date format. Use YYYY-MM-DD, 'today', or no argument for yesterday."
exit 1
fi
else
TARGET_DATE=$(date -v-1d +%Y-%m-%d)
fi
log "========================================"
log "Screenpipe sync starting for: $TARGET_DATE"
log "========================================"
# ─── PREFLIGHT ────────────────────────────────────────────────────────────────
step "Preflight checks"
if [ ! -f "$DB_SRC" ]; then
log "ERROR: Source DB not found at $DB_SRC"; exit 1
fi
printf " %-20s %s (%s)\n" "Source DB:" "OK" "$(du -sh "$DB_SRC" | cut -f1)"
if [ ! -d "$NAS_MOUNT" ]; then
log "ERROR: NAS not mounted at $NAS_MOUNT"; exit 1
fi
printf " %-20s %s\n" "NAS mount:" "OK $NAS_MOUNT"
# Check if DB already synced for this date
DB_ALREADY_SYNCED=false
if [ -f "$NAS_DB" ]; then
EXISTING=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';" 2>/dev/null || echo "0")
if [ "$EXISTING" -gt "0" ]; then
log "Date $TARGET_DATE already has $EXISTING frames in archive — skipping DB sync"
DB_ALREADY_SYNCED=true
else
printf " %-20s %s (%s)\n" "Archive DB:" "exists" "$(du -sh "$NAS_DB" | cut -f1)"
fi
else
printf " %-20s %s\n" "Archive DB:" "will be created"
fi
# Source data dir for this date
DATA_SRC="$HOME/.screenpipe/data/data/$TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
DATA_SIZE=$(du -sh "$DATA_SRC" | cut -f1)
DATA_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
printf " %-20s %s (%s files, %s)\n" "Data dir:" "OK" "$DATA_FILES" "$DATA_SIZE"
else
printf " %-20s %s\n" "Data dir:" "not found — skipping file copy"
fi
# ─── DB SYNC ──────────────────────────────────────────────────────────────────
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── COUNT SOURCE ROWS ────────────────────────────────────────────────────
step "Counting source rows for $TARGET_DATE"
SRC_FRAMES=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
SRC_ELEMENTS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_UI=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
SRC_OCR=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_MEETINGS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start) = '$TARGET_DATE';")
printf " %-20s %s\n" "frames:" "$SRC_FRAMES"
printf " %-20s %s\n" "elements:" "$SRC_ELEMENTS"
printf " %-20s %s\n" "ui_events:" "$SRC_UI"
printf " %-20s %s\n" "ocr_text:" "$SRC_OCR"
printf " %-20s %s\n" "meetings:" "$SRC_MEETINGS"
if [ "$SRC_FRAMES" -eq "0" ]; then
log "No frames found for $TARGET_DATE — skipping DB sync"
DB_ALREADY_SYNCED=true
fi
fi
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── INIT TABLES ──────────────────────────────────────────────────────────
step "Initialising tables, indexes, FTS"
run_sqlite_heredoc "creating tables" "
ATTACH '$NAS_DB' AS nas;
CREATE TABLE IF NOT EXISTS nas.frames AS SELECT * FROM main.frames WHERE 0;
CREATE TABLE IF NOT EXISTS nas.elements AS SELECT * FROM main.elements WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ui_events AS SELECT * FROM main.ui_events WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ocr_text AS SELECT * FROM main.ocr_text WHERE 0;
CREATE TABLE IF NOT EXISTS nas.video_chunks AS SELECT * FROM main.video_chunks WHERE 0;
CREATE TABLE IF NOT EXISTS nas.meetings AS SELECT * FROM main.meetings WHERE 0;
DETACH nas;
"
run_sqlite_heredoc "creating indexes" "
ATTACH '$NAS_DB' AS nas;
CREATE INDEX IF NOT EXISTS nas.idx_frames_timestamp ON frames(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_frames_app_name ON frames(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_window_name ON frames(window_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_video_chunk_id ON frames(video_chunk_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_id ON elements(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_src_role ON elements(frame_id, source, role) WHERE text IS NOT NULL;
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_timestamp ON ui_events(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_app_name ON ui_events(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_frame_id ON ui_events(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_ocr_text_frame_id ON ocr_text(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_meetings_start ON meetings(meeting_start);
CREATE INDEX IF NOT EXISTS nas.idx_video_chunks_device ON video_chunks(device_name);
DETACH nas;
"
run_sqlite_heredoc "creating FTS tables" "
ATTACH '$NAS_DB' AS nas;
CREATE VIRTUAL TABLE IF NOT EXISTS nas.elements_fts USING fts5(
text, role, frame_id UNINDEXED,
content='elements', content_rowid='id', tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.frames_fts USING fts5(
full_text, app_name, window_name, browser_url, id UNINDEXED,
tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.ui_events_fts USING fts5(
text_content,
app_name,
window_title,
element_name,
content='ui_events',
content_rowid='id',
tokenize='unicode61'
);
DETACH nas;
"
# ─── SYNC DATA ────────────────────────────────────────────────────────────
step "Syncing data for $TARGET_DATE"
run_sqlite_heredoc "video_chunks" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.video_chunks
SELECT * FROM main.video_chunks
WHERE id IN (
SELECT DISTINCT video_chunk_id FROM main.frames
WHERE date(timestamp) = '$TARGET_DATE' AND video_chunk_id IS NOT NULL
);
DETACH nas;
"
run_sqlite_heredoc "frames ($SRC_FRAMES rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.frames
SELECT * FROM main.frames WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ocr_text ($SRC_OCR rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ocr_text
SELECT o.* FROM main.ocr_text o
JOIN main.frames f ON o.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ui_events ($SRC_UI rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ui_events
SELECT * FROM main.ui_events WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "elements ($SRC_ELEMENTS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.elements
SELECT e.* FROM main.elements e
JOIN main.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "meetings ($SRC_MEETINGS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.meetings
SELECT * FROM main.meetings WHERE date(meeting_start) = '$TARGET_DATE';
DETACH nas;
"
# ─── FTS UPDATE ───────────────────────────────────────────────────────────
step "Updating FTS indexes"
run_sqlite_heredoc "elements_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.elements_fts(rowid, text, role)
SELECT e.id, e.text, e.role
FROM nas.elements e
JOIN nas.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE'
AND e.text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "frames_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.frames_fts(rowid, full_text, app_name, window_name, browser_url, id)
SELECT id, full_text, app_name, window_name, browser_url, id
FROM nas.frames
WHERE date(timestamp) = '$TARGET_DATE'
AND full_text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "ui_events_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.ui_events_fts(rowid, text_content, app_name, window_title, element_name)
SELECT id, text_content, app_name, window_title, element_name
FROM nas.ui_events
WHERE date(timestamp) = '$TARGET_DATE'
AND text_content IS NOT NULL;
DETACH nas;
"
# ─── VERIFY DB ────────────────────────────────────────────────────────────
step "Verifying DB"
V_FRAMES=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
V_ELEMENTS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_UI=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
V_OCR=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_MEETINGS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start)= '$TARGET_DATE';")
check "frames" "$V_FRAMES" "$SRC_FRAMES"
check "elements" "$V_ELEMENTS" "$SRC_ELEMENTS"
check "ui_events" "$V_UI" "$SRC_UI"
check "ocr_text" "$V_OCR" "$SRC_OCR"
check "meetings" "$V_MEETINGS" "$SRC_MEETINGS"
fi
# ─── COPY DATA FOLDER ─────────────────────────────────────────────────────────
# Always runs regardless of DB sync status
step "Copying data folder for $TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
mkdir -p "$NAS_DATA/$TARGET_DATE"
RSYNC_START=$(date +%s)
printf " %-36s " "rsync $TARGET_DATE/ → NAS"
rsync -a --ignore-existing \
"$DATA_SRC/" \
"$NAS_DATA/$TARGET_DATE/" \
2>>"$LOG_FILE"
RSYNC_DUR=$(( $(date +%s) - RSYNC_START ))
COPIED_FILES=$(ls "$NAS_DATA/$TARGET_DATE" | wc -l | tr -d ' ')
SRC_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
COPIED_SIZE=$(du -sh "$NAS_DATA/$TARGET_DATE" | cut -f1)
if [ "$COPIED_FILES" -eq "$SRC_FILES" ]; then
printf "\r %-36s ✓ %dm%02ds (%s files, %s)\n" \
"rsync $TARGET_DATE/ → NAS" \
"$(( RSYNC_DUR / 60 ))" "$(( RSYNC_DUR % 60 ))" \
"$COPIED_FILES" "$COPIED_SIZE" | tee -a "$LOG_FILE"
else
printf "\r %-36s ✗ %s / %s files\n" \
"rsync $TARGET_DATE/ → NAS" "$COPIED_FILES" "$SRC_FILES" | tee -a "$LOG_FILE"
fi
else
printf " %-36s %s\n" "rsync $TARGET_DATE/ → NAS" "skipped (no source dir)"
fi
# ─── SUMMARY ──────────────────────────────────────────────────────────────────
TOTAL_ELAPSED=$(( $(date +%s) - SCRIPT_START ))
DB_SIZE=$(du -sh "$NAS_DB" | cut -f1)
echo ""
log "Archive DB size: $DB_SIZE"
log "Total time: $(( TOTAL_ELAPSED / 60 ))m$(( TOTAL_ELAPSED % 60 ))s"
log "Sync complete for $TARGET_DATE"
log "========================================"
#!/bin/bash
# screenpipe_sync.sh
# Syncs Screenpipe SQLite data to a NAS archive database (append-only, no deletions).
# Also copies the day's video/frame data folder to the NAS.
#
# Usage:
# ./screenpipe_sync.sh # syncs yesterday (default)
# ./screenpipe_sync.sh 2026-04-15 # syncs a specific date
# ./screenpipe_sync.sh today # syncs today so far
#
# Cron example (runs at 3am daily):
# 0 3 * * * /Users/lukas/.screenpipe/screenpipe_sync.sh >> /Users/lukas/.screenpipe/sync.log 2>&1
set -euo pipefail
# ─── CONFIG ───────────────────────────────────────────────────────────────────
DB_SRC="${SCREENPIPE_DB:-$HOME/.screenpipe/db.sqlite}"
NAS_MOUNT="${NAS_MOUNT:-/Volumes/Test/screenpipe}"
NAS_DB="$NAS_MOUNT/archive.db"
NAS_DATA="$NAS_MOUNT/data"
LOG_FILE="$HOME/.screenpipe/sync.log"
# ──────────────────────────────────────────────────────────────────────────────
# ─── HELPERS ──────────────────────────────────────────────────────────────────
SCRIPT_START=$(date +%s)
log() {
local msg="[$(date '+%Y-%m-%d %H:%M:%S')] $*"
echo "$msg" | tee -a "$LOG_FILE"
}
step() {
local now=$(date +%s)
local elapsed=$(( now - SCRIPT_START ))
local min=$(( elapsed / 60 ))
local sec=$(( elapsed % 60 ))
printf "\n[+%02dm%02ds] ▶ %s\n" "$min" "$sec" "$*" | tee -a "$LOG_FILE"
}
run_sqlite_heredoc() {
local label="$1"
local sql="$2"
local start=$(date +%s)
printf " %-36s " "$label"
sqlite3 "$DB_SRC" <<< "$sql" &
local pid=$!
local spin=[PASSWORD] '⠙' '⠹' '⠸' '⠼' '⠴' '⠦' '⠧' '⠇' '⠏')
local i=0
while kill -0 "$pid" 2>/dev/null; do
printf "\r %-36s %s " "$label" "${spin[$i]}"
i=$(( (i + 1) % 10 ))
sleep 0.2
done
wait "$pid"
local rc=$?
if [ $rc -ne 0 ]; then
printf "\r %-36s ✗ FAILED\n" "$label" | tee -a "$LOG_FILE"
exit $rc
fi
local dur=$(( $(date +%s) - start ))
printf "\r %-36s ✓ %dm%02ds\n" "$label" "$(( dur / 60 ))" "$(( dur % 60 ))" | tee -a "$LOG_FILE"
}
check() {
local label="$1" got="$2" expected="$3"
if [ "$got" -eq "$expected" ]; then
printf " %-20s %s / %s ✓\n" "$label:" "$got" "$expected"
else
printf " %-20s %s / %s ✗ MISMATCH\n" "$label:" "$got" "$expected"
fi
}
# ──────────────────────────────────────────────────────────────────────────────
# ─── DATE ARGUMENT ────────────────────────────────────────────────────────────
if [ "${1:-}" = "today" ]; then
TARGET_DATE=$(date +%Y-%m-%d)
elif [ -n "${1:-}" ]; then
TARGET_DATE="$1"
if ! [[ "$TARGET_DATE" =~ ^[0-9]{4}-[0-9]{2}-[0-9]{2}$ ]]; then
echo "ERROR: Invalid date format. Use YYYY-MM-DD, 'today', or no argument for yesterday."
exit 1
fi
else
TARGET_DATE=$(date -v-1d +%Y-%m-%d)
fi
log "========================================"
log "Screenpipe sync starting for: $TARGET_DATE"
log "========================================"
# ─── PREFLIGHT ────────────────────────────────────────────────────────────────
step "Preflight checks"
if [ ! -f "$DB_SRC" ]; then
log "ERROR: Source DB not found at $DB_SRC"; exit 1
fi
printf " %-20s %s (%s)\n" "Source DB:" "OK" "$(du -sh "$DB_SRC" | cut -f1)"
if [ ! -d "$NAS_MOUNT" ]; then
log "ERROR: NAS not mounted at $NAS_MOUNT"; exit 1
fi
printf " %-20s %s\n" "NAS mount:" "OK $NAS_MOUNT"
# Check if DB already synced for this date
DB_ALREADY_SYNCED=false
if [ -f "$NAS_DB" ]; then
EXISTING=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';" 2>/dev/null || echo "0")
if [ "$EXISTING" -gt "0" ]; then
log "Date $TARGET_DATE already has $EXISTING frames in archive — skipping DB sync"
DB_ALREADY_SYNCED=true
else
printf " %-20s %s (%s)\n" "Archive DB:" "exists" "$(du -sh "$NAS_DB" | cut -f1)"
fi
else
printf " %-20s %s\n" "Archive DB:" "will be created"
fi
# Source data dir for this date
DATA_SRC="$HOME/.screenpipe/data/data/$TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
DATA_SIZE=$(du -sh "$DATA_SRC" | cut -f1)
DATA_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
printf " %-20s %s (%s files, %s)\n" "Data dir:" "OK" "$DATA_FILES" "$DATA_SIZE"
else
printf " %-20s %s\n" "Data dir:" "not found — skipping file copy"
fi
# ─── DB SYNC ──────────────────────────────────────────────────────────────────
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── COUNT SOURCE ROWS ────────────────────────────────────────────────────
step "Counting source rows for $TARGET_DATE"
SRC_FRAMES=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
SRC_ELEMENTS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_UI=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
SRC_OCR=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_MEETINGS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start) = '$TARGET_DATE';")
printf " %-20s %s\n" "frames:" "$SRC_FRAMES"
printf " %-20s %s\n" "elements:" "$SRC_ELEMENTS"
printf " %-20s %s\n" "ui_events:" "$SRC_UI"
printf " %-20s %s\n" "ocr_text:" "$SRC_OCR"
printf " %-20s %s\n" "meetings:" "$SRC_MEETINGS"
if [ "$SRC_FRAMES" -eq "0" ]; then
log "No frames found for $TARGET_DATE — skipping DB sync"
DB_ALREADY_SYNCED=true
fi
fi
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── INIT TABLES ──────────────────────────────────────────────────────────
step "Initialising tables, indexes, FTS"
run_sqlite_heredoc "creating tables" "
ATTACH '$NAS_DB' AS nas;
CREATE TABLE IF NOT EXISTS nas.frames AS SELECT * FROM main.frames WHERE 0;
CREATE TABLE IF NOT EXISTS nas.elements AS SELECT * FROM main.elements WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ui_events AS SELECT * FROM main.ui_events WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ocr_text AS SELECT * FROM main.ocr_text WHERE 0;
CREATE TABLE IF NOT EXISTS nas.video_chunks AS SELECT * FROM main.video_chunks WHERE 0;
CREATE TABLE IF NOT EXISTS nas.meetings AS SELECT * FROM main.meetings WHERE 0;
DETACH nas;
"
run_sqlite_heredoc "creating indexes" "
ATTACH '$NAS_DB' AS nas;
CREATE INDEX IF NOT EXISTS nas.idx_frames_timestamp ON frames(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_frames_app_name ON frames(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_window_name ON frames(window_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_video_chunk_id ON frames(video_chunk_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_id ON elements(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_src_role ON elements(frame_id, source, role) WHERE text IS NOT NULL;
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_timestamp ON ui_events(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_app_name ON ui_events(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_frame_id ON ui_events(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_ocr_text_frame_id ON ocr_text(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_meetings_start ON meetings(meeting_start);
CREATE INDEX IF NOT EXISTS nas.idx_video_chunks_device ON video_chunks(device_name);
DETACH nas;
"
run_sqlite_heredoc "creating FTS tables" "
ATTACH '$NAS_DB' AS nas;
CREATE VIRTUAL TABLE IF NOT EXISTS nas.elements_fts USING fts5(
text, role, frame_id UNINDEXED,
content='elements', content_rowid='id', tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.frames_fts USING fts5(
full_text, app_name, window_name, browser_url, id UNINDEXED,
tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.ui_events_fts USING fts5(
text_content,
app_name,
window_title,
element_name,
content='ui_events',
content_rowid='id',
tokenize='unicode61'
);
DETACH nas;
"
# ─── SYNC DATA ────────────────────────────────────────────────────────────
step "Syncing data for $TARGET_DATE"
run_sqlite_heredoc "video_chunks" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.video_chunks
SELECT * FROM main.video_chunks
WHERE id IN (
SELECT DISTINCT video_chunk_id FROM main.frames
WHERE date(timestamp) = '$TARGET_DATE' AND video_chunk_id IS NOT NULL
);
DETACH nas;
"
run_sqlite_heredoc "frames ($SRC_FRAMES rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.frames
SELECT * FROM main.frames WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ocr_text ($SRC_OCR rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ocr_text
SELECT o.* FROM main.ocr_text o
JOIN main.frames f ON o.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ui_events ($SRC_UI rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ui_events
SELECT * FROM main.ui_events WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "elements ($SRC_ELEMENTS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.elements
SELECT e.* FROM main.elements e
JOIN main.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "meetings ($SRC_MEETINGS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.meetings
SELECT * FROM main.meetings WHERE date(meeting_start) = '$TARGET_DATE';
DETACH nas;
"
# ─── FTS UPDATE ───────────────────────────────────────────────────────────
step "Updating FTS indexes"
run_sqlite_heredoc "elements_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.elements_fts(rowid, text, role)
SELECT e.id, e.text, e.role
FROM nas.elements e
JOIN nas.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE'
AND e.text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "frames_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.frames_fts(rowid, full_text, app_name, window_name, browser_url, id)
SELECT id, full_text, app_name, window_name, browser_url, id
FROM nas.frames
WHERE date(timestamp) = '$TARGET_DATE'
AND full_text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "ui_events_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.ui_events_fts(rowid, text_content, app_name, window_title, element_name)
SELECT id, text_content, app_name, window_title, element_name
FROM nas.ui_events
WHERE date(timestamp) = '$TARGET_DATE'
AND text_content IS NOT NULL;
DETACH nas;
"
# ─── VERIFY DB ────────────────────────────────────────────────────────────
step "Verifying DB"
V_FRAMES=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
V_ELEMENTS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_UI=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
V_OCR=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_MEETINGS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start)= '$TARGET_DATE';")
check "frames" "$V_FRAMES" "$SRC_FRAMES"
check "elements" "$V_ELEMENTS" "$SRC_ELEMENTS"
check "ui_events" "$V_UI" "$SRC_UI"
check "ocr_text" "$V_OCR" "$SRC_OCR"
check "meetings" "$V_MEETINGS" "$SRC_MEETINGS"
fi
# ─── COPY DATA FOLDER ─────────────────────────────────────────────────────────
# Always runs regardless of DB sync status
step "Copying data folder for $TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
mkdir -p "$NAS_DATA/$TARGET_DATE"
RSYNC_START=$(date +%s)
printf " %-36s " "rsync $TARGET_DATE/ → NAS"
rsync -a --ignore-existing \
"$DATA_SRC/" \
"$NAS_DATA/$TARGET_DATE/" \
2>>"$LOG_FILE"
RSYNC_DUR=$(( $(date +%s) - RSYNC_START ))
COPIED_FILES=$(ls "$NAS_DATA/$TARGET_DATE" | wc -l | tr -d ' ')
SRC_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
COPIED_SIZE=$(du -sh "$NAS_DATA/$TARGET_DATE" | cut -f1)
if [ "$COPIED_FILES" -eq "$SRC_FILES" ]; then
printf "\r %-36s ✓ %dm%02ds (%s files, %s)\n" \
"rsync $TARGET_DATE/ → NAS" \
"$(( RSYNC_DUR / 60 ))" "$(( RSYNC_DUR % 60 ))" \
"$COPIED_FILES" "$COPIED_SIZE" | tee -a "$LOG_FILE"
else
printf "\r %-36s ✗ %s / %s files\n" \
"rsync $TARGET_DATE/ → NAS" "$COPIED_FILES" "$SRC_FILES" | tee -a "$LOG_FILE"
fi
else
printf " %-36s %s\n" "rsync $TARGET_DATE/ → NAS" "skipped (no source dir)"
fi
# ─── SUMMARY ──────────────────────────────────────────────────────────────────
TOTAL_ELAPSED=$(( $(date +%s) - SCRIPT_START ))
DB_SIZE=$(du -sh "$NAS_DB" | cut -f1)
echo ""
log "Archive DB size: $DB_SIZE"
log "Total time: $(( TOTAL_ELAPSED / 60 ))m$(( TOTAL_ELAPSED % 60 ))s"
log "Sync complete for $TARGET_DATE"
log "========================================"
lets add ability to jump…, Editor Group 2
Problems (⇧⌘M)
PROBLEMS
Output (⇧⌘U)
OUTPUT
Debug Console (⇧⌘Y)
DEBUG CONSOLE
Terminal (⌃`)
TERMINAL
Ports
PORTS
remote SSH: nas
SSH: nas
screenpipe (Git) - master*, Checkout Branch/Tag...
master*
screenpipe (Git) - Synchronize Changes
No Problems
0
0
No Ports Forwarded
0
Notifications
Signed out
Signed out
Shell Script
Editor Language Status: $(copilot) No inline suggestion available, Inline suggestions
LF
UTF-8
Spaces: 2
Ln 331, Col 48
Screen Reader Optimized
git-commit Lukas Kovalik (23 hours ago)
Lukas Kovalik (23 hours ago)
Info: Downloading VS Code Server...
lets add ability to jump fro,m search to timeline so we can see the exact screen (also based on monitor and time
Session history
New session
Message actions
This session is being continued from a previous conversation that ran out of context. The summary below covers the earlier portion of the conversation.
Summary:
1. Primary Request and Intent:
The user has been developing a Screenpipe dashboard on a Synology NAS. Requests in chronological order:
- **Q1**: Are db.sqlite-shm and db.sqlite-wal used? → Explained WAL mode; cleaned up files.
- **Q2**: AND combination search (term1 + term2 must both appear) → Added "AND also" input field.
- **Q3**: Explain UTC time issue, show date in results, make Only/Skip apps use multiselect → Implemented all three.
- **Q4**: Point sqlite-web at archive.db, add second sqlite-web for db.sqlite → Updated docker-compose.yml, added two services on ports 8767/8768.
- **Q5**: "still the same" (container not picking up config) → Explained `up -d` vs `restart`.
- **Q6**: Add git, push to gitea.com/lakylak/screenpipe.git → Initialized repo, committed, pushed.
- **Q7**: Add Timetable tab with video playback and timeline → Full implementation with click-to-seek, auto-advancing chunks, legend.
- **Q8**: Add scroll/zoom on timeline, jump to specific time, better awareness → Implemented zoom (1×–32×), scrollable timeline, overview minimap, viewport indicator, Jump-to-time input, screenshot thumbnail.
- **Q9 (most recent)**: Make playing speed more than 8×, fix speed not working, show position on timeline → CSS fix done (wider playhead + time label style), remainder in progress.
2. Key Technical Concepts:
- **SQLite WAL mode**: db.sqlite-shm and db.sqlite-wal are WAL companion files; cleaned with `PRAGMA wal_checkpoint(TRUNCATE)`.
- **SQLite FTS5**: Space-separated terms = AND; `MATCH 'term1 term2'` requires both; OR requires explicit `OR` keyword.
- **FastAPI**: Python backend with route handlers in main.py, consumers pattern for DB queries.
- **Docker Compose**: App image is built (code baked in), not volume-mounted → requires `build` + `up -d` for code changes. `restart` doesn't re-read compose file — need `up -d`.
- **sqlite-web**: Read-only flag `-r`, pre-built image, no rebuild needed for config changes.
- **Video playback**: screenpipe records at 0.1–0.3 fps (timelapse). `seek = offset_index / fps` gives seconds into video chunk. ~245 chunks/day across 2 monitors. Videos may be encoded at standard fps making them very short at native speed.
- **Timeline rendering**: App-colored segments built by merging consecutive frames with same `app_name`. Percentage positions: `((ts - dayStart) / span * 100)%`.
- **Zoom/scroll**: Inner div width = `zoom × 100%`, outer wrapper has `overflow-x:auto`. Playhead position stays as % of inner div, which is correct at all zoom levels.
- **Timer-based speed**: For speed > 2×, browser `playbackRate` is unreliable on low-FPS video; correct approach is `setInterval` advancing `video.currentTime` by `speed * interval_ms / 1000` seconds per tick.
- **UTC vs local time**: Backend treats `time_from`/`time_to` as UTC. Frontend converts with `localTimeToUtc()` before sending.
- **Git**: `#recycle/` in .gitignore treated as comment — must escape as `\#recycle/`.
3. Files and Code Sections:
- **`/volume1/screenpipe/app/main.py`**
- Added `/api/apps` endpoint (all unique app names from frames table)
- Added `/api/timetable` endpoint (frames + video URLs + screenshot URLs per device per date)
- `/api/timetable` response structure: `{frames: [{ts, app, win, video_url, seek, screenshot_url}], devices: [...], device: "..."}`
- Key snippet:
```python
@app.get("/api/timetable"
)
async def timetable_data(d: str | None = Query(None), device: str | None = Query(None)):
...
rows = conn.execute(f"""
SELECT f.timestamp AS ts, f.app_name AS app, f.window_name AS win,
f.offset_index, f.snapshot_path,
vc.file_path AS video_path, vc.fps AS video_fps
FROM frames f
JOIN video_chunks vc ON f.video_chunk_id = vc.id
WHERE f.timestamp >= ? AND f.timestamp < ?
AND f.video_chunk_id IS NOT NULL {dev_clause}
ORDER BY f.timestamp
""", params).fetchall()
# seek = offset_index / fps, screenshot_url from _snapshot_to_url(snapshot_path)
```
- **`/volume1/screenpipe/app/docker-compose.yml`**
- Changed sqlite-web service name to `sqlite-web-archive` → port 8767 → `/data/archive.db`
- Added `sqlite-web-live` → port 8768 → `/data/db.sqlite`
- screenpipe-app: port 8766, `DB_PATH: /data/db.sqlite`, `TZ: Europe/Sofia`
- **`/volume1/screenpipe/app/static/index.html`** (single large file with all CSS/HTML/JS)
- **Search enhancements**:
- Added `id="sq2"` "AND also" input between search-row1 and search-row2
- Replaced `sOnlyApps`/`sSkipApps` text inputs with custom multiselect components (`id="msOnlyApps"`, `id="msSkipApps"`)
- Updated `buildUrl()` to combine `q + " " + q2` and convert times with `localTimeToUtc()`
- Added `fmtDateTime(ts)` showing "Apr 20 14:08" format
- Added `localTimeToUtc(timeStr)` converting local HH:MM to UTC
- Multiselect state: `const _msState = { msOnlyApps: new Set(), msSkipApps: new Set() }`
- **Timetable tab** (new page):
- Tab button added between Work Report and AI Summary
- `showPage` array updated: `['activity','search','audio','report','timetable','summary']`
- `refresh()` calls `loadTimetable()` when on timetable page
- **Timetable CSS** (key styles):
```css
.tt-scroll-wrap { overflow-x:auto; ... scrollbar-width:thin; }
.tt-inner { position:relative; height:52px; } /* width set dynamically */
.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
.tt-overview-wrap { position:relative; height:14px; ... cursor:pointer; }
.tt-overview-vp { position:absolute; background:rgba(255,255,255,0.12); border:1px solid rgba(255,255,255,0.4); }
```
- **Timetable JS state variables**:
```javascript
let _ttFrames = [], _ttChunks = [], _ttCurChunk = -1;
let _ttDayStart = 0, _ttDayEnd = 0, _ttZoom = 1;
const _ttAppColors = {};
const TT_ZOOMS = [1, 2, 4, 8, 16, 32];
const TT_TICK_MIN = {1:30, 2:15, 4:10, 8:5, 16:2, 32:1};
```
- **Key JS functions**:
- `loadTimetable()` — fetches `/api/timetable`, groups chunks, renders
- `ttZoom(dir)` — changes zoom level, updates inner width, re-renders ticks, scrolls to playhead
- `ttTimelineClick(e)` — uses `inner.getBoundingClientRect()` (works correctly when scrolled)
- `ttOverviewClick(e)` — scrolls timeline + seeks video
- `ttUpdateOverviewVp()` — updates white viewport rect using `wrap.scrollLeft / inner.scrollWidth`
- `ttJumpToTime()` — parses time input, converts to ms, calls `ttSeekToMs`
- `ttSeekToMs(tsMs)` — binary-searches nearest frame, loads chunk, seeks video
- `_ttScrollToPlayhead(force)` — scrolls wrap so playhead stays at ~35% from left when Follow is checked
- `_ttMovePlayhead(tsMs)` — sets `ph.style.left` as percentage, calls `_ttScrollToPlayhead`
- `_ttShowNow(f)` — updates screenshot thumbnail + now-text div
- `ttSyncBtn()` — syncs play/pause button text
- Keyboard: Space=play/pause, ←/→=±10s, Ctrl+←/→=±30s, +/-=zoom
- **`/volume1/screenpipe/.gitignore`**
```
*.db, *.sqlite, *.sqlite-shm, *.sqlite-wal
data/, \#recycle/,
@eaDir/
app_settings.json, app/.env
__pycache__/, *.pyc, .DS_Store
```
4. Errors and Fixes:
- **`#recycle/` in .gitignore treated as comment**: `#` starts a comment in .gitignore. Fixed by escaping: `\#recycle/`. Also had to `git rm --cached "#recycle/desktop.ini"` to unstage it.
- **`sudo docker compose restart sqlite-web` didn't apply config change**: `restart` reuses existing container config, doesn't re-read compose file. Fix: `sudo docker compose up -d` recreates containers with new config.
- **Orphan container holding port 8767**: Old `screenpipe-sqlweb` container still running after rename. Fix: `sudo docker compose up -d --remove-orphans`.
- **`sudo docker` not accessible from Claude Code**: Interactive sudo required. User runs commands in NAS terminal.
- **`git config --global --add safe.directory /volume1/screenpipe`**: Required due to dubious ownership of NAS volume directory.
5. Problem Solving:
- **Time zone confusion in search**: Backend uses UTC for time_from/time_to; frontend displays local time. Solved by `localTimeToUtc()` converting before API call, and showing date+time in results.
- **sqlite-web not showing April 17–21 data**: Container was pointing at db.sqlite (old config), not archive.db. Fixed by updating docker-compose.yml + `up -d`.
- **Video playback speed**: Browser `video.playbackRate` is capped (~16× max in Chrome) and performs poorly for very low FPS (0.1–0.3 fps) screenpipe videos. Each chunk may only be 1–2 seconds of wall-clock encoded video, playing through instantly. Solution: implement timer-based frame-jumping for speed > 2× (in progress).
- **Playhead visibility**: White 2px line was hard to see. Fixed CSS to 3px width + stronger box-shadow + added `.tt-playhead-time` overlay label style.
6. All User Messages:
- "do we use db.sqlite-shm and db.sqlite-wal for anything right now? [image of search UI] Lets add ability to search with combination... Explain how does the search work time seems to work in utc maybe display date next to the time in results. Make only apps and skip apps multiselect with all available apps."
- "doi I need to restart?"
- "there is sudo docker compose"
- "no still the same"
- "add git. I have created gitea repo https://gitea.com/lakylak/screenpipe.git"
- "just do it" (re: git push)
- "how to make sure data in :8767 are accurate and point to archive.db"
- "I can't see older than 16th there should 21st of april. In http://[IP_ADDRESS]:8766/ I can see all but not on sqlite viewer http://[IP_ADDRESS]:8767"
- "ok lets add additional port to read db.sqlite, I would like to see the structure of official db that reads db.sqlite db and one for archive"
- [docker compose output showing orphan container error]
- "looks great. I want to have ability to scroll left and right and also to jump to specific time. I want to be more aware to what I see. Still it should be fast"
- "lets make the playing speed more than 8x"
- "still playing speed is not showing properly still just in one second. also I want to see in timeline where I am"
7. Pending Tasks:
- **Playback speed fix** (in progress, partially done):
- Add `<div class="tt-playhead-time" id="ttPlayheadTime"></div>` inside `ttInner` HTML
- Implement `_ttStartSpeedTimer(speed)` / `_ttStopSpeedTimer()` / `ttSetSpeed(speed)` functions
- Update speed select `onchange` to call `ttSetSpeed(+this.value)`
- Add 64× option to speed select
- Update `_ttMovePlayhead` to also update `ttPlayheadTime` label
- Update `ttSyncBtn` to start/stop timer based on play state + speed
- Update `ttNextChunk` and `ttSeekToMs` to stop timer (restarted by onplay → ttSyncBtn)
- Rebuild Docker image after changes
8. Current Work:
Working on fixing two issues with the Timetable tab:
1. **Playback speed**: `video.playbackRate` doesn't work well for low-FPS screenpipe videos. The plan is to implement a `setInterval`-based frame-jumper for speed > 2× that advances `video.currentTime` by `speed * intervalMs / 1000` every 100–200ms, bypassing browser playbackRate limits.
2. **Timeline position indicator**: Playhead not prominent enough.
**CSS already changed** in `/volume1/screenpipe/app/static/index.html`:
```css
.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
```
Session was interrupted by summary request after this CSS edit, mid-implementation.
9. Optional Next Step:
Continuing the fix for: "still playing speed is not showing properly still just in one second. also I want to see in timeline where I am"
The next steps in order are:
1. Add `<div class="tt-playhead-time" id="ttPlayheadTime"></div>` to the HTML inside `#ttInner` (after the existing `.tt-playhead` div)
2. Update `_ttMovePlayhead` to set `ttPlayheadTime` position and text content with `fmtTime()`
3. Add timer-based speed system to JS (new functions `_ttStartSpeedTimer`, `_ttStopSpeedTimer`, `ttSetSpeed`)
4. Update speed select's `onchange` to `ttSetSpeed(+this.value)` and add 64× option
5. Update `ttSyncBtn`, `ttNextChunk`, `ttSeekToMs` to integrate with timer
6. Rebuild: `sudo docker compose build screenpipe-app && sudo docker compose up -d screenpipe-app`
If you need specific details from before compaction (like exact code snippets, error messages, or content you generated), read the full transcript at: /home/Adm1n/.claude/projects/-volume1-screenpipe/7777f46e-1e40-4051-b4bb-f43aedd59556.jsonl
Continue the conversation from where it left off without asking the user any further questions. Resume directly — do not acknowledge the summary, do not recap what was happening, do not preface with "I'll continue" or similar. Pick up the last task as if the break never happened.
Thinking
Thinking
Bash
Find timetable speed and playhead related code
IN
grep -n "tt-playhead\|ttPlayhead\|ttInner\|_ttMovePlayhead\|ttSetSpeed\|_ttStartSpeed\|_ttStopSpeed\|ttSyncBtn\|playbackRate\|speed.*select\|spdSel\|ttSpeed" /volume1/screenpipe/app/static/index.html | head -80
Copy code to clipboard
OUT
207:.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
208:.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
580: <div class="tt-inner" id="ttInner" onclick="ttTimelineClick(event)">
583: <div class="tt-playhead" id="ttPlayhead"></div>
598: onpause="ttSyncBtn()"
599: onplay="ttSyncBtn()">
607: <select id="ttSpeed" onchange="document.getElementById('ttVideo').playbackRate=+this.value">
1624: document.getElementById('ttInner').style.width = '100%';
1632: document.getElementById('ttPlayhead').style.display = 'none';
1694: document.getElementById('ttInner').style.width = (_ttZoom * 100) + '%';
1700: const inner = document.getElementById('ttInner');
1710: const inner = document.getElementById('ttInner');
1718: const inner = document.getElementById('ttInner');
1750: _ttMovePlayhead(new Date(best.ts).getTime());
1772: _ttMovePlayhead(new Date(best.ts).getTime());
1791:function _ttMovePlayhead(tsMs) {
1792: const ph = document.getElementById('ttPlayhead');
1802: const ph = document.getElementById('ttPlayhead');
1805: const inner = document.getElementById('ttInner');
1819:function ttSyncBtn() {
Read
index.html
index.html
(lines 576-625)
Read
index.html
index.html
(lines 1786-1865)
Thinking
Thinking
Bash
Find chunk loading and playhead update functions
IN
grep -n "ttUpdatePlayhead\|ttNextChunk\|ttLoadChunk\|_ttCurChunk\|_ttChunks\|_ttFrames" /volume1/screenpipe/app/static/index.html | head -50
Copy code to clipboard
OUT
596: onended="ttNextChunk()"
597: ontimeupdate="ttUpdatePlayhead()"
1569:let _ttFrames = [];
1570:let _ttChunks = [];
1571:let _ttCurChunk = -1;
1607: _ttFrames = data.frames;
1609: for (const f of _ttFrames) {
1613: _ttChunks = [...chunkMap.values()];
1614: _ttCurChunk = -1;
1616: _ttDayStart = new Date(_ttFrames[0].ts).getTime();
1617: _ttDayEnd = new Date(_ttFrames[_ttFrames.length - 1].ts).getTime();
1620: for (const f of _ttFrames) _ttColor(f.app);
1640: for (const f of _ttFrames) {
1736: if (!_ttFrames.length) return;
1737: let best = _ttFrames[0], minDiff = Infinity;
1738: for (const f of _ttFrames) {
1742: const ci = _ttChunks.findIndex(ch => ch[0].video_url === best.video_url);
1744: _ttCurChunk = ci;
1753:function ttNextChunk() {
1754: if (_ttCurChunk + 1 >= _ttChunks.length) return;
1755: _ttCurChunk++;
1757: video.src = _ttChunks[_ttCurChunk][0].video_url;
1762:function ttUpdatePlayhead() {
1764: if (_ttCurChunk < 0 || _ttCurChunk >= _ttChunks.length) return;
1765: const chunk = _ttChunks[_ttCurChunk];
Read
index.html
index.html
(lines 1754-1793)
Thinking
Thinking
Read
index.html
index.html
(lines 1701-1760)
Thinking
Thinking
Read
index.html
index.html
(lines 1561-1640)
Thinking
Thinking
Bash
Find time formatting functions
IN
grep -n "function fmtTime\|function fmtDate\|function _fmt\|localTimeToUtc\|fmtDateTime" /volume1/screenpipe/app/static/index.html | head -20
Copy code to clipboard
OUT
1088: if (tf) url += `&time_from=${localTimeToUtc(tf)}`;
1089: if (tt) url += `&time_to=${localTimeToUtc(tt)}`;
1153: let meta = `<span style="color:var(--muted)">${fmtDateTime(r.timestamp)}</span>${srcBadge}`;
1166: meta = `<span style="color:var(--muted)">${fmtDateTime(r.timestamp)}</span>${srcBadge} <span class="badge ${r.is_input_device?'mic':'sys'}">${r.is_input_device?'MIC':'SYS'}</span>`;
1307:function fmtTime(ts) {
1312:function fmtDateTime(ts) {
1320:function localTimeToUtc(timeStr) {
1778:function _fmtSec(s) {
1788: `${fmtDateTime(f.ts)} · ${f.app}${f.win ? ' / ' + f.win : ''}`;
Read
index.html
index.html
(lines 1308-1327)
Thinking
Thinking
Now I have everything I need. I'll implement the frame-by-frame timer system that replaces
video.playbackRate
with wall-clock accurate playback, plus add the playhead time label. Let me make the changes:
Edit
index.html
index.html
Added 1 line...
|
Code
|
screenpipe_sync.sh — screenpipe [SSH: nas]
|
NULL
|
|
Explorer (⇧⌘E)
Search (⇧⌘F)
Source Control (⌃⇧ Explorer (⇧⌘E)
Search (⇧⌘F)
Source Control (⌃⇧G) - 2 pending changes
2
Run and Debug (⇧⌘D)
Remote Explorer
Extensions (⇧⌘X) - 2 require update, 1 requires restart
3
Claude Code
Containers
EXPLORER
EXPLORER
Explorer Section: screenpipe [SSH: nas]
Explorer Section: screenpipe [SSH: nas]
SCREENPIPE [SSH: NAS]
#recycle
app
data
pipes
.gitignore
app_settings.json
archive.db
db.sqlite
screenpipe_sync.sh
screenpipe.db
Outline Section
OUTLINE
OUTLINE
Timeline Section
TIMELINE
TIMELINE
app_settings.json, Editor Group 1
screenpipe_sync.sh, Editor Group 1
db.sqlite-shm, preview, Editor Group 1
#!/bin/bash
# screenpipe_sync.sh
# Syncs Screenpipe SQLite data to a NAS archive database (append-only, no deletions).
# Also copies the day's video/frame data folder to the NAS.
#
# Usage:
# ./screenpipe_sync.sh # syncs yesterday (default)
# ./screenpipe_sync.sh 2026-04-15 # syncs a specific date
# ./screenpipe_sync.sh today # syncs today so far
#
# Cron example (runs at 3am daily):
# 0 3 * * * /Users/lukas/.screenpipe/screenpipe_sync.sh >> /Users/lukas/.screenpipe/sync.log 2>&1
set -euo pipefail
# ─── CONFIG ───────────────────────────────────────────────────────────────────
DB_SRC="${SCREENPIPE_DB:-$HOME/.screenpipe/db.sqlite}"
NAS_MOUNT="${NAS_MOUNT:-/Volumes/Test/screenpipe}"
NAS_DB="$NAS_MOUNT/archive.db"
NAS_DATA="$NAS_MOUNT/data"
LOG_FILE="$HOME/.screenpipe/sync.log"
# ──────────────────────────────────────────────────────────────────────────────
# ─── HELPERS ──────────────────────────────────────────────────────────────────
SCRIPT_START=$(date +%s)
log() {
local msg="[$(date '+%Y-%m-%d %H:%M:%S')] $*"
echo "$msg" | tee -a "$LOG_FILE"
}
step() {
local now=$(date +%s)
local elapsed=$(( now - SCRIPT_START ))
local min=$(( elapsed / 60 ))
local sec=$(( elapsed % 60 ))
printf "\n[+%02dm%02ds] ▶ %s\n" "$min" "$sec" "$*" | tee -a "$LOG_FILE"
}
run_sqlite_heredoc() {
local label="$1"
local sql="$2"
local start=$(date +%s)
printf " %-36s " "$label"
sqlite3 "$DB_SRC" <<< "$sql" &
local pid=$!
local spin=[PASSWORD] '⠙' '⠹' '⠸' '⠼' '⠴' '⠦' '⠧' '⠇' '⠏')
local i=0
while kill -0 "$pid" 2>/dev/null; do
printf "\r %-36s %s " "$label" "${spin[$i]}"
i=$(( (i + 1) % 10 ))
sleep 0.2
done
wait "$pid"
local rc=$?
if [ $rc -ne 0 ]; then
printf "\r %-36s ✗ FAILED\n" "$label" | tee -a "$LOG_FILE"
exit $rc
fi
local dur=$(( $(date +%s) - start ))
printf "\r %-36s ✓ %dm%02ds\n" "$label" "$(( dur / 60 ))" "$(( dur % 60 ))" | tee -a "$LOG_FILE"
}
check() {
local label="$1" got="$2" expected="$3"
if [ "$got" -eq "$expected" ]; then
printf " %-20s %s / %s ✓\n" "$label:" "$got" "$expected"
else
printf " %-20s %s / %s ✗ MISMATCH\n" "$label:" "$got" "$expected"
fi
}
# ──────────────────────────────────────────────────────────────────────────────
# ─── DATE ARGUMENT ────────────────────────────────────────────────────────────
if [ "${1:-}" = "today" ]; then
TARGET_DATE=$(date +%Y-%m-%d)
elif [ -n "${1:-}" ]; then
TARGET_DATE="$1"
if ! [[ "$TARGET_DATE" =~ ^[0-9]{4}-[0-9]{2}-[0-9]{2}$ ]]; then
echo "ERROR: Invalid date format. Use YYYY-MM-DD, 'today', or no argument for yesterday."
exit 1
fi
else
TARGET_DATE=$(date -v-1d +%Y-%m-%d)
fi
log "========================================"
log "Screenpipe sync starting for: $TARGET_DATE"
log "========================================"
# ─── PREFLIGHT ────────────────────────────────────────────────────────────────
step "Preflight checks"
if [ ! -f "$DB_SRC" ]; then
log "ERROR: Source DB not found at $DB_SRC"; exit 1
fi
printf " %-20s %s (%s)\n" "Source DB:" "OK" "$(du -sh "$DB_SRC" | cut -f1)"
if [ ! -d "$NAS_MOUNT" ]; then
log "ERROR: NAS not mounted at $NAS_MOUNT"; exit 1
fi
printf " %-20s %s\n" "NAS mount:" "OK $NAS_MOUNT"
# Check if DB already synced for this date
DB_ALREADY_SYNCED=false
if [ -f "$NAS_DB" ]; then
EXISTING=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';" 2>/dev/null || echo "0")
if [ "$EXISTING" -gt "0" ]; then
log "Date $TARGET_DATE already has $EXISTING frames in archive — skipping DB sync"
DB_ALREADY_SYNCED=true
else
printf " %-20s %s (%s)\n" "Archive DB:" "exists" "$(du -sh "$NAS_DB" | cut -f1)"
fi
else
printf " %-20s %s\n" "Archive DB:" "will be created"
fi
# Source data dir for this date
DATA_SRC="$HOME/.screenpipe/data/data/$TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
DATA_SIZE=$(du -sh "$DATA_SRC" | cut -f1)
DATA_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
printf " %-20s %s (%s files, %s)\n" "Data dir:" "OK" "$DATA_FILES" "$DATA_SIZE"
else
printf " %-20s %s\n" "Data dir:" "not found — skipping file copy"
fi
# ─── DB SYNC ──────────────────────────────────────────────────────────────────
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── COUNT SOURCE ROWS ────────────────────────────────────────────────────
step "Counting source rows for $TARGET_DATE"
SRC_FRAMES=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
SRC_ELEMENTS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_UI=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
SRC_OCR=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_MEETINGS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start) = '$TARGET_DATE';")
printf " %-20s %s\n" "frames:" "$SRC_FRAMES"
printf " %-20s %s\n" "elements:" "$SRC_ELEMENTS"
printf " %-20s %s\n" "ui_events:" "$SRC_UI"
printf " %-20s %s\n" "ocr_text:" "$SRC_OCR"
printf " %-20s %s\n" "meetings:" "$SRC_MEETINGS"
if [ "$SRC_FRAMES" -eq "0" ]; then
log "No frames found for $TARGET_DATE — skipping DB sync"
DB_ALREADY_SYNCED=true
fi
fi
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── INIT TABLES ──────────────────────────────────────────────────────────
step "Initialising tables, indexes, FTS"
run_sqlite_heredoc "creating tables" "
ATTACH '$NAS_DB' AS nas;
CREATE TABLE IF NOT EXISTS nas.frames AS SELECT * FROM main.frames WHERE 0;
CREATE TABLE IF NOT EXISTS nas.elements AS SELECT * FROM main.elements WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ui_events AS SELECT * FROM main.ui_events WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ocr_text AS SELECT * FROM main.ocr_text WHERE 0;
CREATE TABLE IF NOT EXISTS nas.video_chunks AS SELECT * FROM main.video_chunks WHERE 0;
CREATE TABLE IF NOT EXISTS nas.meetings AS SELECT * FROM main.meetings WHERE 0;
DETACH nas;
"
run_sqlite_heredoc "creating indexes" "
ATTACH '$NAS_DB' AS nas;
CREATE INDEX IF NOT EXISTS nas.idx_frames_timestamp ON frames(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_frames_app_name ON frames(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_window_name ON frames(window_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_video_chunk_id ON frames(video_chunk_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_id ON elements(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_src_role ON elements(frame_id, source, role) WHERE text IS NOT NULL;
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_timestamp ON ui_events(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_app_name ON ui_events(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_frame_id ON ui_events(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_ocr_text_frame_id ON ocr_text(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_meetings_start ON meetings(meeting_start);
CREATE INDEX IF NOT EXISTS nas.idx_video_chunks_device ON video_chunks(device_name);
DETACH nas;
"
run_sqlite_heredoc "creating FTS tables" "
ATTACH '$NAS_DB' AS nas;
CREATE VIRTUAL TABLE IF NOT EXISTS nas.elements_fts USING fts5(
text, role, frame_id UNINDEXED,
content='elements', content_rowid='id', tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.frames_fts USING fts5(
full_text, app_name, window_name, browser_url, id UNINDEXED,
tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.ui_events_fts USING fts5(
text_content,
app_name,
window_title,
element_name,
content='ui_events',
content_rowid='id',
tokenize='unicode61'
);
DETACH nas;
"
# ─── SYNC DATA ────────────────────────────────────────────────────────────
step "Syncing data for $TARGET_DATE"
run_sqlite_heredoc "video_chunks" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.video_chunks
SELECT * FROM main.video_chunks
WHERE id IN (
SELECT DISTINCT video_chunk_id FROM main.frames
WHERE date(timestamp) = '$TARGET_DATE' AND video_chunk_id IS NOT NULL
);
DETACH nas;
"
run_sqlite_heredoc "frames ($SRC_FRAMES rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.frames
SELECT * FROM main.frames WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ocr_text ($SRC_OCR rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ocr_text
SELECT o.* FROM main.ocr_text o
JOIN main.frames f ON o.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ui_events ($SRC_UI rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ui_events
SELECT * FROM main.ui_events WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "elements ($SRC_ELEMENTS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.elements
SELECT e.* FROM main.elements e
JOIN main.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "meetings ($SRC_MEETINGS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.meetings
SELECT * FROM main.meetings WHERE date(meeting_start) = '$TARGET_DATE';
DETACH nas;
"
# ─── FTS UPDATE ───────────────────────────────────────────────────────────
step "Updating FTS indexes"
run_sqlite_heredoc "elements_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.elements_fts(rowid, text, role)
SELECT e.id, e.text, e.role
FROM nas.elements e
JOIN nas.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE'
AND e.text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "frames_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.frames_fts(rowid, full_text, app_name, window_name, browser_url, id)
SELECT id, full_text, app_name, window_name, browser_url, id
FROM nas.frames
WHERE date(timestamp) = '$TARGET_DATE'
AND full_text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "ui_events_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.ui_events_fts(rowid, text_content, app_name, window_title, element_name)
SELECT id, text_content, app_name, window_title, element_name
FROM nas.ui_events
WHERE date(timestamp) = '$TARGET_DATE'
AND text_content IS NOT NULL;
DETACH nas;
"
# ─── VERIFY DB ────────────────────────────────────────────────────────────
step "Verifying DB"
V_FRAMES=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
V_ELEMENTS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_UI=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
V_OCR=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_MEETINGS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start)= '$TARGET_DATE';")
check "frames" "$V_FRAMES" "$SRC_FRAMES"
check "elements" "$V_ELEMENTS" "$SRC_ELEMENTS"
check "ui_events" "$V_UI" "$SRC_UI"
check "ocr_text" "$V_OCR" "$SRC_OCR"
check "meetings" "$V_MEETINGS" "$SRC_MEETINGS"
fi
# ─── COPY DATA FOLDER ─────────────────────────────────────────────────────────
# Always runs regardless of DB sync status
step "Copying data folder for $TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
mkdir -p "$NAS_DATA/$TARGET_DATE"
RSYNC_START=$(date +%s)
printf " %-36s " "rsync $TARGET_DATE/ → NAS"
rsync -a --ignore-existing \
"$DATA_SRC/" \
"$NAS_DATA/$TARGET_DATE/" \
2>>"$LOG_FILE"
RSYNC_DUR=$(( $(date +%s) - RSYNC_START ))
COPIED_FILES=$(ls "$NAS_DATA/$TARGET_DATE" | wc -l | tr -d ' ')
SRC_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
COPIED_SIZE=$(du -sh "$NAS_DATA/$TARGET_DATE" | cut -f1)
if [ "$COPIED_FILES" -eq "$SRC_FILES" ]; then
printf "\r %-36s ✓ %dm%02ds (%s files, %s)\n" \
"rsync $TARGET_DATE/ → NAS" \
"$(( RSYNC_DUR / 60 ))" "$(( RSYNC_DUR % 60 ))" \
"$COPIED_FILES" "$COPIED_SIZE" | tee -a "$LOG_FILE"
else
printf "\r %-36s ✗ %s / %s files\n" \
"rsync $TARGET_DATE/ → NAS" "$COPIED_FILES" "$SRC_FILES" | tee -a "$LOG_FILE"
fi
else
printf " %-36s %s\n" "rsync $TARGET_DATE/ → NAS" "skipped (no source dir)"
fi
# ─── SUMMARY ──────────────────────────────────────────────────────────────────
TOTAL_ELAPSED=$(( $(date +%s) - SCRIPT_START ))
DB_SIZE=$(du -sh "$NAS_DB" | cut -f1)
echo ""
log "Archive DB size: $DB_SIZE"
log "Total time: $(( TOTAL_ELAPSED / 60 ))m$(( TOTAL_ELAPSED % 60 ))s"
log "Sync complete for $TARGET_DATE"
log "========================================"
#!/bin/bash
# screenpipe_sync.sh
# Syncs Screenpipe SQLite data to a NAS archive database (append-only, no deletions).
# Also copies the day's video/frame data folder to the NAS.
#
# Usage:
# ./screenpipe_sync.sh # syncs yesterday (default)
# ./screenpipe_sync.sh 2026-04-15 # syncs a specific date
# ./screenpipe_sync.sh today # syncs today so far
#
# Cron example (runs at 3am daily):
# 0 3 * * * /Users/lukas/.screenpipe/screenpipe_sync.sh >> /Users/lukas/.screenpipe/sync.log 2>&1
set -euo pipefail
# ─── CONFIG ───────────────────────────────────────────────────────────────────
DB_SRC="${SCREENPIPE_DB:-$HOME/.screenpipe/db.sqlite}"
NAS_MOUNT="${NAS_MOUNT:-/Volumes/Test/screenpipe}"
NAS_DB="$NAS_MOUNT/archive.db"
NAS_DATA="$NAS_MOUNT/data"
LOG_FILE="$HOME/.screenpipe/sync.log"
# ──────────────────────────────────────────────────────────────────────────────
# ─── HELPERS ──────────────────────────────────────────────────────────────────
SCRIPT_START=$(date +%s)
log() {
local msg="[$(date '+%Y-%m-%d %H:%M:%S')] $*"
echo "$msg" | tee -a "$LOG_FILE"
}
step() {
local now=$(date +%s)
local elapsed=$(( now - SCRIPT_START ))
local min=$(( elapsed / 60 ))
local sec=$(( elapsed % 60 ))
printf "\n[+%02dm%02ds] ▶ %s\n" "$min" "$sec" "$*" | tee -a "$LOG_FILE"
}
run_sqlite_heredoc() {
local label="$1"
local sql="$2"
local start=$(date +%s)
printf " %-36s " "$label"
sqlite3 "$DB_SRC" <<< "$sql" &
local pid=$!
local spin=[PASSWORD] '⠙' '⠹' '⠸' '⠼' '⠴' '⠦' '⠧' '⠇' '⠏')
local i=0
while kill -0 "$pid" 2>/dev/null; do
printf "\r %-36s %s " "$label" "${spin[$i]}"
i=$(( (i + 1) % 10 ))
sleep 0.2
done
wait "$pid"
local rc=$?
if [ $rc -ne 0 ]; then
printf "\r %-36s ✗ FAILED\n" "$label" | tee -a "$LOG_FILE"
exit $rc
fi
local dur=$(( $(date +%s) - start ))
printf "\r %-36s ✓ %dm%02ds\n" "$label" "$(( dur / 60 ))" "$(( dur % 60 ))" | tee -a "$LOG_FILE"
}
check() {
local label="$1" got="$2" expected="$3"
if [ "$got" -eq "$expected" ]; then
printf " %-20s %s / %s ✓\n" "$label:" "$got" "$expected"
else
printf " %-20s %s / %s ✗ MISMATCH\n" "$label:" "$got" "$expected"
fi
}
# ──────────────────────────────────────────────────────────────────────────────
# ─── DATE ARGUMENT ────────────────────────────────────────────────────────────
if [ "${1:-}" = "today" ]; then
TARGET_DATE=$(date +%Y-%m-%d)
elif [ -n "${1:-}" ]; then
TARGET_DATE="$1"
if ! [[ "$TARGET_DATE" =~ ^[0-9]{4}-[0-9]{2}-[0-9]{2}$ ]]; then
echo "ERROR: Invalid date format. Use YYYY-MM-DD, 'today', or no argument for yesterday."
exit 1
fi
else
TARGET_DATE=$(date -v-1d +%Y-%m-%d)
fi
log "========================================"
log "Screenpipe sync starting for: $TARGET_DATE"
log "========================================"
# ─── PREFLIGHT ────────────────────────────────────────────────────────────────
step "Preflight checks"
if [ ! -f "$DB_SRC" ]; then
log "ERROR: Source DB not found at $DB_SRC"; exit 1
fi
printf " %-20s %s (%s)\n" "Source DB:" "OK" "$(du -sh "$DB_SRC" | cut -f1)"
if [ ! -d "$NAS_MOUNT" ]; then
log "ERROR: NAS not mounted at $NAS_MOUNT"; exit 1
fi
printf " %-20s %s\n" "NAS mount:" "OK $NAS_MOUNT"
# Check if DB already synced for this date
DB_ALREADY_SYNCED=false
if [ -f "$NAS_DB" ]; then
EXISTING=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';" 2>/dev/null || echo "0")
if [ "$EXISTING" -gt "0" ]; then
log "Date $TARGET_DATE already has $EXISTING frames in archive — skipping DB sync"
DB_ALREADY_SYNCED=true
else
printf " %-20s %s (%s)\n" "Archive DB:" "exists" "$(du -sh "$NAS_DB" | cut -f1)"
fi
else
printf " %-20s %s\n" "Archive DB:" "will be created"
fi
# Source data dir for this date
DATA_SRC="$HOME/.screenpipe/data/data/$TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
DATA_SIZE=$(du -sh "$DATA_SRC" | cut -f1)
DATA_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
printf " %-20s %s (%s files, %s)\n" "Data dir:" "OK" "$DATA_FILES" "$DATA_SIZE"
else
printf " %-20s %s\n" "Data dir:" "not found — skipping file copy"
fi
# ─── DB SYNC ──────────────────────────────────────────────────────────────────
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── COUNT SOURCE ROWS ────────────────────────────────────────────────────
step "Counting source rows for $TARGET_DATE"
SRC_FRAMES=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
SRC_ELEMENTS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_UI=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
SRC_OCR=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_MEETINGS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start) = '$TARGET_DATE';")
printf " %-20s %s\n" "frames:" "$SRC_FRAMES"
printf " %-20s %s\n" "elements:" "$SRC_ELEMENTS"
printf " %-20s %s\n" "ui_events:" "$SRC_UI"
printf " %-20s %s\n" "ocr_text:" "$SRC_OCR"
printf " %-20s %s\n" "meetings:" "$SRC_MEETINGS"
if [ "$SRC_FRAMES" -eq "0" ]; then
log "No frames found for $TARGET_DATE — skipping DB sync"
DB_ALREADY_SYNCED=true
fi
fi
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── INIT TABLES ──────────────────────────────────────────────────────────
step "Initialising tables, indexes, FTS"
run_sqlite_heredoc "creating tables" "
ATTACH '$NAS_DB' AS nas;
CREATE TABLE IF NOT EXISTS nas.frames AS SELECT * FROM main.frames WHERE 0;
CREATE TABLE IF NOT EXISTS nas.elements AS SELECT * FROM main.elements WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ui_events AS SELECT * FROM main.ui_events WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ocr_text AS SELECT * FROM main.ocr_text WHERE 0;
CREATE TABLE IF NOT EXISTS nas.video_chunks AS SELECT * FROM main.video_chunks WHERE 0;
CREATE TABLE IF NOT EXISTS nas.meetings AS SELECT * FROM main.meetings WHERE 0;
DETACH nas;
"
run_sqlite_heredoc "creating indexes" "
ATTACH '$NAS_DB' AS nas;
CREATE INDEX IF NOT EXISTS nas.idx_frames_timestamp ON frames(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_frames_app_name ON frames(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_window_name ON frames(window_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_video_chunk_id ON frames(video_chunk_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_id ON elements(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_src_role ON elements(frame_id, source, role) WHERE text IS NOT NULL;
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_timestamp ON ui_events(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_app_name ON ui_events(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_frame_id ON ui_events(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_ocr_text_frame_id ON ocr_text(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_meetings_start ON meetings(meeting_start);
CREATE INDEX IF NOT EXISTS nas.idx_video_chunks_device ON video_chunks(device_name);
DETACH nas;
"
run_sqlite_heredoc "creating FTS tables" "
ATTACH '$NAS_DB' AS nas;
CREATE VIRTUAL TABLE IF NOT EXISTS nas.elements_fts USING fts5(
text, role, frame_id UNINDEXED,
content='elements', content_rowid='id', tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.frames_fts USING fts5(
full_text, app_name, window_name, browser_url, id UNINDEXED,
tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.ui_events_fts USING fts5(
text_content,
app_name,
window_title,
element_name,
content='ui_events',
content_rowid='id',
tokenize='unicode61'
);
DETACH nas;
"
# ─── SYNC DATA ────────────────────────────────────────────────────────────
step "Syncing data for $TARGET_DATE"
run_sqlite_heredoc "video_chunks" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.video_chunks
SELECT * FROM main.video_chunks
WHERE id IN (
SELECT DISTINCT video_chunk_id FROM main.frames
WHERE date(timestamp) = '$TARGET_DATE' AND video_chunk_id IS NOT NULL
);
DETACH nas;
"
run_sqlite_heredoc "frames ($SRC_FRAMES rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.frames
SELECT * FROM main.frames WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ocr_text ($SRC_OCR rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ocr_text
SELECT o.* FROM main.ocr_text o
JOIN main.frames f ON o.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ui_events ($SRC_UI rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ui_events
SELECT * FROM main.ui_events WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "elements ($SRC_ELEMENTS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.elements
SELECT e.* FROM main.elements e
JOIN main.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "meetings ($SRC_MEETINGS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.meetings
SELECT * FROM main.meetings WHERE date(meeting_start) = '$TARGET_DATE';
DETACH nas;
"
# ─── FTS UPDATE ───────────────────────────────────────────────────────────
step "Updating FTS indexes"
run_sqlite_heredoc "elements_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.elements_fts(rowid, text, role)
SELECT e.id, e.text, e.role
FROM nas.elements e
JOIN nas.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE'
AND e.text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "frames_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.frames_fts(rowid, full_text, app_name, window_name, browser_url, id)
SELECT id, full_text, app_name, window_name, browser_url, id
FROM nas.frames
WHERE date(timestamp) = '$TARGET_DATE'
AND full_text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "ui_events_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.ui_events_fts(rowid, text_content, app_name, window_title, element_name)
SELECT id, text_content, app_name, window_title, element_name
FROM nas.ui_events
WHERE date(timestamp) = '$TARGET_DATE'
AND text_content IS NOT NULL;
DETACH nas;
"
# ─── VERIFY DB ────────────────────────────────────────────────────────────
step "Verifying DB"
V_FRAMES=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
V_ELEMENTS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_UI=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
V_OCR=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_MEETINGS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start)= '$TARGET_DATE';")
check "frames" "$V_FRAMES" "$SRC_FRAMES"
check "elements" "$V_ELEMENTS" "$SRC_ELEMENTS"
check "ui_events" "$V_UI" "$SRC_UI"
check "ocr_text" "$V_OCR" "$SRC_OCR"
check "meetings" "$V_MEETINGS" "$SRC_MEETINGS"
fi
# ─── COPY DATA FOLDER ─────────────────────────────────────────────────────────
# Always runs regardless of DB sync status
step "Copying data folder for $TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
mkdir -p "$NAS_DATA/$TARGET_DATE"
RSYNC_START=$(date +%s)
printf " %-36s " "rsync $TARGET_DATE/ → NAS"
rsync -a --ignore-existing \
"$DATA_SRC/" \
"$NAS_DATA/$TARGET_DATE/" \
2>>"$LOG_FILE"
RSYNC_DUR=$(( $(date +%s) - RSYNC_START ))
COPIED_FILES=$(ls "$NAS_DATA/$TARGET_DATE" | wc -l | tr -d ' ')
SRC_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
COPIED_SIZE=$(du -sh "$NAS_DATA/$TARGET_DATE" | cut -f1)
if [ "$COPIED_FILES" -eq "$SRC_FILES" ]; then
printf "\r %-36s ✓ %dm%02ds (%s files, %s)\n" \
"rsync $TARGET_DATE/ → NAS" \
"$(( RSYNC_DUR / 60 ))" "$(( RSYNC_DUR % 60 ))" \
"$COPIED_FILES" "$COPIED_SIZE" | tee -a "$LOG_FILE"
else
printf "\r %-36s ✗ %s / %s files\n" \
"rsync $TARGET_DATE/ → NAS" "$COPIED_FILES" "$SRC_FILES" | tee -a "$LOG_FILE"
fi
else
printf " %-36s %s\n" "rsync $TARGET_DATE/ → NAS" "skipped (no source dir)"
fi
# ─── SUMMARY ──────────────────────────────────────────────────────────────────
TOTAL_ELAPSED=$(( $(date +%s) - SCRIPT_START ))
DB_SIZE=$(du -sh "$NAS_DB" | cut -f1)
echo ""
log "Archive DB size: $DB_SIZE"
log "Total time: $(( TOTAL_ELAPSED / 60 ))m$(( TOTAL_ELAPSED % 60 ))s"
log "Sync complete for $TARGET_DATE"
log "========================================"
lets add ability to jump…, Editor Group 2
Problems (⇧⌘M)
PROBLEMS
Output (⇧⌘U)
OUTPUT
Debug Console (⇧⌘Y)
DEBUG CONSOLE
Terminal (⌃`)
TERMINAL
Ports
PORTS
remote SSH: nas
SSH: nas
screenpipe (Git) - master*, Checkout Branch/Tag...
master*
screenpipe (Git) - Synchronize Changes
No Problems
0
0
No Ports Forwarded
0
Notifications
Signed out
Signed out
Shell Script
Editor Language Status: $(copilot) No inline suggestion available, Inline suggestions
LF
UTF-8
Spaces: 2
Ln 331, Col 48
Screen Reader Optimized
git-commit Lukas Kovalik (23 hours ago)
Lukas Kovalik (23 hours ago)
Info: Downloading VS Code Server...
lets add ability to jump fro,m search to timeline so we can see the exact screen (also based on monitor and time
Session history
New session
Message actions
This session is being continued from a previous conversation that ran out of context. The summary below covers the earlier portion of the conversation.
Summary:
1. Primary Request and Intent:
The user has been developing a Screenpipe dashboard on a Synology NAS. Requests in chronological order:
- **Q1**: Are db.sqlite-shm and db.sqlite-wal used? → Explained WAL mode; cleaned up files.
- **Q2**: AND combination search (term1 + term2 must both appear) → Added "AND also" input field.
- **Q3**: Explain UTC time issue, show date in results, make Only/Skip apps use multiselect → Implemented all three.
- **Q4**: Point sqlite-web at archive.db, add second sqlite-web for db.sqlite → Updated docker-compose.yml, added two services on ports 8767/8768.
- **Q5**: "still the same" (container not picking up config) → Explained `up -d` vs `restart`.
- **Q6**: Add git, push to gitea.com/lakylak/screenpipe.git → Initialized repo, committed, pushed.
- **Q7**: Add Timetable tab with video playback and timeline → Full implementation with click-to-seek, auto-advancing chunks, legend.
- **Q8**: Add scroll/zoom on timeline, jump to specific time, better awareness → Implemented zoom (1×–32×), scrollable timeline, overview minimap, viewport indicator, Jump-to-time input, screenshot thumbnail.
- **Q9 (most recent)**: Make playing speed more than 8×, fix speed not working, show position on timeline → CSS fix done (wider playhead + time label style), remainder in progress.
2. Key Technical Concepts:
- **SQLite WAL mode**: db.sqlite-shm and db.sqlite-wal are WAL companion files; cleaned with `PRAGMA wal_checkpoint(TRUNCATE)`.
- **SQLite FTS5**: Space-separated terms = AND; `MATCH 'term1 term2'` requires both; OR requires explicit `OR` keyword.
- **FastAPI**: Python backend with route handlers in main.py, consumers pattern for DB queries.
- **Docker Compose**: App image is built (code baked in), not volume-mounted → requires `build` + `up -d` for code changes. `restart` doesn't re-read compose file — need `up -d`.
- **sqlite-web**: Read-only flag `-r`, pre-built image, no rebuild needed for config changes.
- **Video playback**: screenpipe records at 0.1–0.3 fps (timelapse). `seek = offset_index / fps` gives seconds into video chunk. ~245 chunks/day across 2 monitors. Videos may be encoded at standard fps making them very short at native speed.
- **Timeline rendering**: App-colored segments built by merging consecutive frames with same `app_name`. Percentage positions: `((ts - dayStart) / span * 100)%`.
- **Zoom/scroll**: Inner div width = `zoom × 100%`, outer wrapper has `overflow-x:auto`. Playhead position stays as % of inner div, which is correct at all zoom levels.
- **Timer-based speed**: For speed > 2×, browser `playbackRate` is unreliable on low-FPS video; correct approach is `setInterval` advancing `video.currentTime` by `speed * interval_ms / 1000` seconds per tick.
- **UTC vs local time**: Backend treats `time_from`/`time_to` as UTC. Frontend converts with `localTimeToUtc()` before sending.
- **Git**: `#recycle/` in .gitignore treated as comment — must escape as `\#recycle/`.
3. Files and Code Sections:
- **`/volume1/screenpipe/app/main.py`**
- Added `/api/apps` endpoint (all unique app names from frames table)
- Added `/api/timetable` endpoint (frames + video URLs + screenshot URLs per device per date)
- `/api/timetable` response structure: `{frames: [{ts, app, win, video_url, seek, screenshot_url}], devices: [...], device: "..."}`
- Key snippet:
```python
@app.get("/api/timetable"
)
async def timetable_data(d: str | None = Query(None), device: str | None = Query(None)):
...
rows = conn.execute(f"""
SELECT f.timestamp AS ts, f.app_name AS app, f.window_name AS win,
f.offset_index, f.snapshot_path,
vc.file_path AS video_path, vc.fps AS video_fps
FROM frames f
JOIN video_chunks vc ON f.video_chunk_id = vc.id
WHERE f.timestamp >= ? AND f.timestamp < ?
AND f.video_chunk_id IS NOT NULL {dev_clause}
ORDER BY f.timestamp
""", params).fetchall()
# seek = offset_index / fps, screenshot_url from _snapshot_to_url(snapshot_path)
```
- **`/volume1/screenpipe/app/docker-compose.yml`**
- Changed sqlite-web service name to `sqlite-web-archive` → port 8767 → `/data/archive.db`
- Added `sqlite-web-live` → port 8768 → `/data/db.sqlite`
- screenpipe-app: port 8766, `DB_PATH: /data/db.sqlite`, `TZ: Europe/Sofia`
- **`/volume1/screenpipe/app/static/index.html`** (single large file with all CSS/HTML/JS)
- **Search enhancements**:
- Added `id="sq2"` "AND also" input between search-row1 and search-row2
- Replaced `sOnlyApps`/`sSkipApps` text inputs with custom multiselect components (`id="msOnlyApps"`, `id="msSkipApps"`)
- Updated `buildUrl()` to combine `q + " " + q2` and convert times with `localTimeToUtc()`
- Added `fmtDateTime(ts)` showing "Apr 20 14:08" format
- Added `localTimeToUtc(timeStr)` converting local HH:MM to UTC
- Multiselect state: `const _msState = { msOnlyApps: new Set(), msSkipApps: new Set() }`
- **Timetable tab** (new page):
- Tab button added between Work Report and AI Summary
- `showPage` array updated: `['activity','search','audio','report','timetable','summary']`
- `refresh()` calls `loadTimetable()` when on timetable page
- **Timetable CSS** (key styles):
```css
.tt-scroll-wrap { overflow-x:auto; ... scrollbar-width:thin; }
.tt-inner { position:relative; height:52px; } /* width set dynamically */
.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
.tt-overview-wrap { position:relative; height:14px; ... cursor:pointer; }
.tt-overview-vp { position:absolute; background:rgba(255,255,255,0.12); border:1px solid rgba(255,255,255,0.4); }
```
- **Timetable JS state variables**:
```javascript
let _ttFrames = [], _ttChunks = [], _ttCurChunk = -1;
let _ttDayStart = 0, _ttDayEnd = 0, _ttZoom = 1;
const _ttAppColors = {};
const TT_ZOOMS = [1, 2, 4, 8, 16, 32];
const TT_TICK_MIN = {1:30, 2:15, 4:10, 8:5, 16:2, 32:1};
```
- **Key JS functions**:
- `loadTimetable()` — fetches `/api/timetable`, groups chunks, renders
- `ttZoom(dir)` — changes zoom level, updates inner width, re-renders ticks, scrolls to playhead
- `ttTimelineClick(e)` — uses `inner.getBoundingClientRect()` (works correctly when scrolled)
- `ttOverviewClick(e)` — scrolls timeline + seeks video
- `ttUpdateOverviewVp()` — updates white viewport rect using `wrap.scrollLeft / inner.scrollWidth`
- `ttJumpToTime()` — parses time input, converts to ms, calls `ttSeekToMs`
- `ttSeekToMs(tsMs)` — binary-searches nearest frame, loads chunk, seeks video
- `_ttScrollToPlayhead(force)` — scrolls wrap so playhead stays at ~35% from left when Follow is checked
- `_ttMovePlayhead(tsMs)` — sets `ph.style.left` as percentage, calls `_ttScrollToPlayhead`
- `_ttShowNow(f)` — updates screenshot thumbnail + now-text div
- `ttSyncBtn()` — syncs play/pause button text
- Keyboard: Space=play/pause, ←/→=±10s, Ctrl+←/→=±30s, +/-=zoom
- **`/volume1/screenpipe/.gitignore`**
```
*.db, *.sqlite, *.sqlite-shm, *.sqlite-wal
data/, \#recycle/,
@eaDir/
app_settings.json, app/.env
__pycache__/, *.pyc, .DS_Store
```
4. Errors and Fixes:
- **`#recycle/` in .gitignore treated as comment**: `#` starts a comment in .gitignore. Fixed by escaping: `\#recycle/`. Also had to `git rm --cached "#recycle/desktop.ini"` to unstage it.
- **`sudo docker compose restart sqlite-web` didn't apply config change**: `restart` reuses existing container config, doesn't re-read compose file. Fix: `sudo docker compose up -d` recreates containers with new config.
- **Orphan container holding port 8767**: Old `screenpipe-sqlweb` container still running after rename. Fix: `sudo docker compose up -d --remove-orphans`.
- **`sudo docker` not accessible from Claude Code**: Interactive sudo required. User runs commands in NAS terminal.
- **`git config --global --add safe.directory /volume1/screenpipe`**: Required due to dubious ownership of NAS volume directory.
5. Problem Solving:
- **Time zone confusion in search**: Backend uses UTC for time_from/time_to; frontend displays local time. Solved by `localTimeToUtc()` converting before API call, and showing date+time in results.
- **sqlite-web not showing April 17–21 data**: Container was pointing at db.sqlite (old config), not archive.db. Fixed by updating docker-compose.yml + `up -d`.
- **Video playback speed**: Browser `video.playbackRate` is capped (~16× max in Chrome) and performs poorly for very low FPS (0.1–0.3 fps) screenpipe videos. Each chunk may only be 1–2 seconds of wall-clock encoded video, playing through instantly. Solution: implement timer-based frame-jumping for speed > 2× (in progress).
- **Playhead visibility**: White 2px line was hard to see. Fixed CSS to 3px width + stronger box-shadow + added `.tt-playhead-time` overlay label style.
6. All User Messages:
- "do we use db.sqlite-shm and db.sqlite-wal for anything right now? [image of search UI] Lets add ability to search with combination... Explain how does the search work time seems to work in utc maybe display date next to the time in results. Make only apps and skip apps multiselect with all available apps."
- "doi I need to restart?"
- "there is sudo docker compose"
- "no still the same"
- "add git. I have created gitea repo https://gitea.com/lakylak/screenpipe.git"
- "just do it" (re: git push)
- "how to make sure data in :8767 are accurate and point to archive.db"
- "I can't see older than 16th there should 21st of april. In http://[IP_ADDRESS]:8766/ I can see all but not on sqlite viewer http://[IP_ADDRESS]:8767"
- "ok lets add additional port to read db.sqlite, I would like to see the structure of official db that reads db.sqlite db and one for archive"
- [docker compose output showing orphan container error]
- "looks great. I want to have ability to scroll left and right and also to jump to specific time. I want to be more aware to what I see. Still it should be fast"
- "lets make the playing speed more than 8x"
- "still playing speed is not showing properly still just in one second. also I want to see in timeline where I am"
7. Pending Tasks:
- **Playback speed fix** (in progress, partially done):
- Add `<div class="tt-playhead-time" id="ttPlayheadTime"></div>` inside `ttInner` HTML
- Implement `_ttStartSpeedTimer(speed)` / `_ttStopSpeedTimer()` / `ttSetSpeed(speed)` functions
- Update speed select `onchange` to call `ttSetSpeed(+this.value)`
- Add 64× option to speed select
- Update `_ttMovePlayhead` to also update `ttPlayheadTime` label
- Update `ttSyncBtn` to start/stop timer based on play state + speed
- Update `ttNextChunk` and `ttSeekToMs` to stop timer (restarted by onplay → ttSyncBtn)
- Rebuild Docker image after changes
8. Current Work:
Working on fixing two issues with the Timetable tab:
1. **Playback speed**: `video.playbackRate` doesn't work well for low-FPS screenpipe videos. The plan is to implement a `setInterval`-based frame-jumper for speed > 2× that advances `video.currentTime` by `speed * intervalMs / 1000` every 100–200ms, bypassing browser playbackRate limits.
2. **Timeline position indicator**: Playhead not prominent enough.
**CSS already changed** in `/volume1/screenpipe/app/static/index.html`:
```css
.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
```
Session was interrupted by summary request after this CSS edit, mid-implementation.
9. Optional Next Step:
Continuing the fix for: "still playing speed is not showing properly still just in one second. also I want to see in timeline where I am"
The next steps in order are:
1. Add `<div class="tt-playhead-time" id="ttPlayheadTime"></div>` to the HTML inside `#ttInner` (after the existing `.tt-playhead` div)
2. Update `_ttMovePlayhead` to set `ttPlayheadTime` position and text content with `fmtTime()`
3. Add timer-based speed system to JS (new functions `_ttStartSpeedTimer`, `_ttStopSpeedTimer`, `ttSetSpeed`)
4. Update speed select's `onchange` to `ttSetSpeed(+this.value)` and add 64× option
5. Update `ttSyncBtn`, `ttNextChunk`, `ttSeekToMs` to integrate with timer
6. Rebuild: `sudo docker compose build screenpipe-app && sudo docker compose up -d screenpipe-app`
If you need specific details from before compaction (like exact code snippets, error messages, or content you generated), read the full transcript at: /home/Adm1n/.claude/projects/-volume1-screenpipe/7777f46e-1e40-4051-b4bb-f43aedd59556.jsonl
Continue the conversation from where it left off without asking the user any further questions. Resume directly — do not acknowledge the summary, do not recap what was happening, do not preface with "I'll continue" or similar. Pick up the last task as if the break never happened.
Thinking
Thinking
Bash
Find timetable speed and playhead related code
IN
grep -n "tt-playhead\|ttPlayhead\|ttInner\|_ttMovePlayhead\|ttSetSpeed\|_ttStartSpeed\|_ttStopSpeed\|ttSyncBtn\|playbackRate\|speed.*select\|spdSel\|ttSpeed" /volume1/screenpipe/app/static/index.html | head -80
Copy code to clipboard
OUT
207:.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
208:.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
580: <div class="tt-inner" id="ttInner" onclick="ttTimelineClick(event)">
583: <div class="tt-playhead" id="ttPlayhead"></div>
598: onpause="ttSyncBtn()"
599: onplay="ttSyncBtn()">
607: <select id="ttSpeed" onchange="document.getElementById('ttVideo').playbackRate=+this.value">
1624: document.getElementById('ttInner').style.width = '100%';
1632: document.getElementById('ttPlayhead').style.display = 'none';
1694: document.getElementById('ttInner').style.width = (_ttZoom * 100) + '%';
1700: const inner = document.getElementById('ttInner');
1710: const inner = document.getElementById('ttInner');
1718: const inner = document.getElementById('ttInner');
1750: _ttMovePlayhead(new Date(best.ts).getTime());
1772: _ttMovePlayhead(new Date(best.ts).getTime());
1791:function _ttMovePlayhead(tsMs) {
1792: const ph = document.getElementById('ttPlayhead');
1802: const ph = document.getElementById('ttPlayhead');
1805: const inner = document.getElementById('ttInner');
1819:function ttSyncBtn() {
Read
index.html
index.html
(lines 576-625)
Read
index.html
index.html
(lines 1786-1865)
Thinking
Thinking
Bash
Find chunk loading and playhead update functions
IN
grep -n "ttUpdatePlayhead\|ttNextChunk\|ttLoadChunk\|_ttCurChunk\|_ttChunks\|_ttFrames" /volume1/screenpipe/app/static/index.html | head -50
Copy code to clipboard
OUT
596: onended="ttNextChunk()"
597: ontimeupdate="ttUpdatePlayhead()"
1569:let _ttFrames = [];
1570:let _ttChunks = [];
1571:let _ttCurChunk = -1;
1607: _ttFrames = data.frames;
1609: for (const f of _ttFrames) {
1613: _ttChunks = [...chunkMap.values()];
1614: _ttCurChunk = -1;
1616: _ttDayStart = new Date(_ttFrames[0].ts).getTime();
1617: _ttDayEnd = new Date(_ttFrames[_ttFrames.length - 1].ts).getTime();
1620: for (const f of _ttFrames) _ttColor(f.app);
1640: for (const f of _ttFrames) {
1736: if (!_ttFrames.length) return;
1737: let best = _ttFrames[0], minDiff = Infinity;
1738: for (const f of _ttFrames) {
1742: const ci = _ttChunks.findIndex(ch => ch[0].video_url === best.video_url);
1744: _ttCurChunk = ci;
1753:function ttNextChunk() {
1754: if (_ttCurChunk + 1 >= _ttChunks.length) return;
1755: _ttCurChunk++;
1757: video.src = _ttChunks[_ttCurChunk][0].video_url;
1762:function ttUpdatePlayhead() {
1764: if (_ttCurChunk < 0 || _ttCurChunk >= _ttChunks.length) return;
1765: const chunk = _ttChunks[_ttCurChunk];
Read
index.html
index.html
(lines 1754-1793)
Thinking
Thinking
Read
index.html
index.html
(lines 1701-1760)
Thinking
Thinking
Read
index.html
index.html
(lines 1561-1640)
Thinking
Thinking
Bash
Find time formatting functions
IN
grep -n "function fmtTime\|function fmtDate\|function _fmt\|localTimeToUtc\|fmtDateTime" /volume1/screenpipe/app/static/index.html | head -20
Copy code to clipboard
OUT
1088: if (tf) url += `&time_from=${localTimeToUtc(tf)}`;
1089: if (tt) url += `&time_to=${localTimeToUtc(tt)}`;
1153: let meta = `<span style="color:var(--muted)">${fmtDateTime(r.timestamp)}</span>${srcBadge}`;
1166: meta = `<span style="color:var(--muted)">${fmtDateTime(r.timestamp)}</span>${srcBadge} <span class="badge ${r.is_input_device?'mic':'sys'}">${r.is_input_device?'MIC':'SYS'}</span>`;
1307:function fmtTime(ts) {
1312:function fmtDateTime(ts) {
1320:function localTimeToUtc(timeStr) {
1778:function _fmtSec(s) {
1788: `${fmtDateTime(f.ts)} · ${f.app}${f.win ? ' / ' + f.win : ''}`;
Read
index.html
index.html
(lines 1308-1327)
Thinking
Thinking
Now I have everything I need. I'll implement the frame-by-frame timer system that replaces
video.playbackRate
with wall-clock accurate playback, plus add the playhead time label. Let me make the changes:
Edit
index.html
index.html
Added 1 line...
|
Code
|
screenpipe_sync.sh — screenpipe [SSH: nas]
|
NULL
|
|
Explorer (⇧⌘E)
Search (⇧⌘F)
Source Control (⌃⇧ Explorer (⇧⌘E)
Search (⇧⌘F)
Source Control (⌃⇧G) - 2 pending changes
2
Run and Debug (⇧⌘D)
Remote Explorer
Extensions (⇧⌘X) - 2 require update, 1 requires restart
3
Claude Code
Containers
EXPLORER
EXPLORER
Explorer Section: screenpipe [SSH: nas]
Explorer Section: screenpipe [SSH: nas]
SCREENPIPE [SSH: NAS]
#recycle
app
data
pipes
.gitignore
app_settings.json
archive.db
db.sqlite
screenpipe_sync.sh
screenpipe.db
Outline Section
OUTLINE
OUTLINE
Timeline Section
TIMELINE
TIMELINE
app_settings.json, Editor Group 1
screenpipe_sync.sh, Editor Group 1
db.sqlite-shm, preview, Editor Group 1
#!/bin/bash
# screenpipe_sync.sh
# Syncs Screenpipe SQLite data to a NAS archive database (append-only, no deletions).
# Also copies the day's video/frame data folder to the NAS.
#
# Usage:
# ./screenpipe_sync.sh # syncs yesterday (default)
# ./screenpipe_sync.sh 2026-04-15 # syncs a specific date
# ./screenpipe_sync.sh today # syncs today so far
#
# Cron example (runs at 3am daily):
# 0 3 * * * /Users/lukas/.screenpipe/screenpipe_sync.sh >> /Users/lukas/.screenpipe/sync.log 2>&1
set -euo pipefail
# ─── CONFIG ───────────────────────────────────────────────────────────────────
DB_SRC="${SCREENPIPE_DB:-$HOME/.screenpipe/db.sqlite}"
NAS_MOUNT="${NAS_MOUNT:-/Volumes/Test/screenpipe}"
NAS_DB="$NAS_MOUNT/archive.db"
NAS_DATA="$NAS_MOUNT/data"
LOG_FILE="$HOME/.screenpipe/sync.log"
# ──────────────────────────────────────────────────────────────────────────────
# ─── HELPERS ──────────────────────────────────────────────────────────────────
SCRIPT_START=$(date +%s)
log() {
local msg="[$(date '+%Y-%m-%d %H:%M:%S')] $*"
echo "$msg" | tee -a "$LOG_FILE"
}
step() {
local now=$(date +%s)
local elapsed=$(( now - SCRIPT_START ))
local min=$(( elapsed / 60 ))
local sec=$(( elapsed % 60 ))
printf "\n[+%02dm%02ds] ▶ %s\n" "$min" "$sec" "$*" | tee -a "$LOG_FILE"
}
run_sqlite_heredoc() {
local label="$1"
local sql="$2"
local start=$(date +%s)
printf " %-36s " "$label"
sqlite3 "$DB_SRC" <<< "$sql" &
local pid=$!
local spin=[PASSWORD] '⠙' '⠹' '⠸' '⠼' '⠴' '⠦' '⠧' '⠇' '⠏')
local i=0
while kill -0 "$pid" 2>/dev/null; do
printf "\r %-36s %s " "$label" "${spin[$i]}"
i=$(( (i + 1) % 10 ))
sleep 0.2
done
wait "$pid"
local rc=$?
if [ $rc -ne 0 ]; then
printf "\r %-36s ✗ FAILED\n" "$label" | tee -a "$LOG_FILE"
exit $rc
fi
local dur=$(( $(date +%s) - start ))
printf "\r %-36s ✓ %dm%02ds\n" "$label" "$(( dur / 60 ))" "$(( dur % 60 ))" | tee -a "$LOG_FILE"
}
check() {
local label="$1" got="$2" expected="$3"
if [ "$got" -eq "$expected" ]; then
printf " %-20s %s / %s ✓\n" "$label:" "$got" "$expected"
else
printf " %-20s %s / %s ✗ MISMATCH\n" "$label:" "$got" "$expected"
fi
}
# ──────────────────────────────────────────────────────────────────────────────
# ─── DATE ARGUMENT ────────────────────────────────────────────────────────────
if [ "${1:-}" = "today" ]; then
TARGET_DATE=$(date +%Y-%m-%d)
elif [ -n "${1:-}" ]; then
TARGET_DATE="$1"
if ! [[ "$TARGET_DATE" =~ ^[0-9]{4}-[0-9]{2}-[0-9]{2}$ ]]; then
echo "ERROR: Invalid date format. Use YYYY-MM-DD, 'today', or no argument for yesterday."
exit 1
fi
else
TARGET_DATE=$(date -v-1d +%Y-%m-%d)
fi
log "========================================"
log "Screenpipe sync starting for: $TARGET_DATE"
log "========================================"
# ─── PREFLIGHT ────────────────────────────────────────────────────────────────
step "Preflight checks"
if [ ! -f "$DB_SRC" ]; then
log "ERROR: Source DB not found at $DB_SRC"; exit 1
fi
printf " %-20s %s (%s)\n" "Source DB:" "OK" "$(du -sh "$DB_SRC" | cut -f1)"
if [ ! -d "$NAS_MOUNT" ]; then
log "ERROR: NAS not mounted at $NAS_MOUNT"; exit 1
fi
printf " %-20s %s\n" "NAS mount:" "OK $NAS_MOUNT"
# Check if DB already synced for this date
DB_ALREADY_SYNCED=false
if [ -f "$NAS_DB" ]; then
EXISTING=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';" 2>/dev/null || echo "0")
if [ "$EXISTING" -gt "0" ]; then
log "Date $TARGET_DATE already has $EXISTING frames in archive — skipping DB sync"
DB_ALREADY_SYNCED=true
else
printf " %-20s %s (%s)\n" "Archive DB:" "exists" "$(du -sh "$NAS_DB" | cut -f1)"
fi
else
printf " %-20s %s\n" "Archive DB:" "will be created"
fi
# Source data dir for this date
DATA_SRC="$HOME/.screenpipe/data/data/$TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
DATA_SIZE=$(du -sh "$DATA_SRC" | cut -f1)
DATA_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
printf " %-20s %s (%s files, %s)\n" "Data dir:" "OK" "$DATA_FILES" "$DATA_SIZE"
else
printf " %-20s %s\n" "Data dir:" "not found — skipping file copy"
fi
# ─── DB SYNC ──────────────────────────────────────────────────────────────────
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── COUNT SOURCE ROWS ────────────────────────────────────────────────────
step "Counting source rows for $TARGET_DATE"
SRC_FRAMES=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
SRC_ELEMENTS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_UI=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
SRC_OCR=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_MEETINGS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start) = '$TARGET_DATE';")
printf " %-20s %s\n" "frames:" "$SRC_FRAMES"
printf " %-20s %s\n" "elements:" "$SRC_ELEMENTS"
printf " %-20s %s\n" "ui_events:" "$SRC_UI"
printf " %-20s %s\n" "ocr_text:" "$SRC_OCR"
printf " %-20s %s\n" "meetings:" "$SRC_MEETINGS"
if [ "$SRC_FRAMES" -eq "0" ]; then
log "No frames found for $TARGET_DATE — skipping DB sync"
DB_ALREADY_SYNCED=true
fi
fi
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── INIT TABLES ──────────────────────────────────────────────────────────
step "Initialising tables, indexes, FTS"
run_sqlite_heredoc "creating tables" "
ATTACH '$NAS_DB' AS nas;
CREATE TABLE IF NOT EXISTS nas.frames AS SELECT * FROM main.frames WHERE 0;
CREATE TABLE IF NOT EXISTS nas.elements AS SELECT * FROM main.elements WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ui_events AS SELECT * FROM main.ui_events WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ocr_text AS SELECT * FROM main.ocr_text WHERE 0;
CREATE TABLE IF NOT EXISTS nas.video_chunks AS SELECT * FROM main.video_chunks WHERE 0;
CREATE TABLE IF NOT EXISTS nas.meetings AS SELECT * FROM main.meetings WHERE 0;
DETACH nas;
"
run_sqlite_heredoc "creating indexes" "
ATTACH '$NAS_DB' AS nas;
CREATE INDEX IF NOT EXISTS nas.idx_frames_timestamp ON frames(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_frames_app_name ON frames(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_window_name ON frames(window_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_video_chunk_id ON frames(video_chunk_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_id ON elements(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_src_role ON elements(frame_id, source, role) WHERE text IS NOT NULL;
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_timestamp ON ui_events(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_app_name ON ui_events(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_frame_id ON ui_events(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_ocr_text_frame_id ON ocr_text(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_meetings_start ON meetings(meeting_start);
CREATE INDEX IF NOT EXISTS nas.idx_video_chunks_device ON video_chunks(device_name);
DETACH nas;
"
run_sqlite_heredoc "creating FTS tables" "
ATTACH '$NAS_DB' AS nas;
CREATE VIRTUAL TABLE IF NOT EXISTS nas.elements_fts USING fts5(
text, role, frame_id UNINDEXED,
content='elements', content_rowid='id', tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.frames_fts USING fts5(
full_text, app_name, window_name, browser_url, id UNINDEXED,
tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.ui_events_fts USING fts5(
text_content,
app_name,
window_title,
element_name,
content='ui_events',
content_rowid='id',
tokenize='unicode61'
);
DETACH nas;
"
# ─── SYNC DATA ────────────────────────────────────────────────────────────
step "Syncing data for $TARGET_DATE"
run_sqlite_heredoc "video_chunks" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.video_chunks
SELECT * FROM main.video_chunks
WHERE id IN (
SELECT DISTINCT video_chunk_id FROM main.frames
WHERE date(timestamp) = '$TARGET_DATE' AND video_chunk_id IS NOT NULL
);
DETACH nas;
"
run_sqlite_heredoc "frames ($SRC_FRAMES rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.frames
SELECT * FROM main.frames WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ocr_text ($SRC_OCR rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ocr_text
SELECT o.* FROM main.ocr_text o
JOIN main.frames f ON o.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ui_events ($SRC_UI rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ui_events
SELECT * FROM main.ui_events WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "elements ($SRC_ELEMENTS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.elements
SELECT e.* FROM main.elements e
JOIN main.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "meetings ($SRC_MEETINGS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.meetings
SELECT * FROM main.meetings WHERE date(meeting_start) = '$TARGET_DATE';
DETACH nas;
"
# ─── FTS UPDATE ───────────────────────────────────────────────────────────
step "Updating FTS indexes"
run_sqlite_heredoc "elements_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.elements_fts(rowid, text, role)
SELECT e.id, e.text, e.role
FROM nas.elements e
JOIN nas.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE'
AND e.text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "frames_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.frames_fts(rowid, full_text, app_name, window_name, browser_url, id)
SELECT id, full_text, app_name, window_name, browser_url, id
FROM nas.frames
WHERE date(timestamp) = '$TARGET_DATE'
AND full_text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "ui_events_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.ui_events_fts(rowid, text_content, app_name, window_title, element_name)
SELECT id, text_content, app_name, window_title, element_name
FROM nas.ui_events
WHERE date(timestamp) = '$TARGET_DATE'
AND text_content IS NOT NULL;
DETACH nas;
"
# ─── VERIFY DB ────────────────────────────────────────────────────────────
step "Verifying DB"
V_FRAMES=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
V_ELEMENTS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_UI=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
V_OCR=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_MEETINGS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start)= '$TARGET_DATE';")
check "frames" "$V_FRAMES" "$SRC_FRAMES"
check "elements" "$V_ELEMENTS" "$SRC_ELEMENTS"
check "ui_events" "$V_UI" "$SRC_UI"
check "ocr_text" "$V_OCR" "$SRC_OCR"
check "meetings" "$V_MEETINGS" "$SRC_MEETINGS"
fi
# ─── COPY DATA FOLDER ─────────────────────────────────────────────────────────
# Always runs regardless of DB sync status
step "Copying data folder for $TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
mkdir -p "$NAS_DATA/$TARGET_DATE"
RSYNC_START=$(date +%s)
printf " %-36s " "rsync $TARGET_DATE/ → NAS"
rsync -a --ignore-existing \
"$DATA_SRC/" \
"$NAS_DATA/$TARGET_DATE/" \
2>>"$LOG_FILE"
RSYNC_DUR=$(( $(date +%s) - RSYNC_START ))
COPIED_FILES=$(ls "$NAS_DATA/$TARGET_DATE" | wc -l | tr -d ' ')
SRC_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
COPIED_SIZE=$(du -sh "$NAS_DATA/$TARGET_DATE" | cut -f1)
if [ "$COPIED_FILES" -eq "$SRC_FILES" ]; then
printf "\r %-36s ✓ %dm%02ds (%s files, %s)\n" \
"rsync $TARGET_DATE/ → NAS" \
"$(( RSYNC_DUR / 60 ))" "$(( RSYNC_DUR % 60 ))" \
"$COPIED_FILES" "$COPIED_SIZE" | tee -a "$LOG_FILE"
else
printf "\r %-36s ✗ %s / %s files\n" \
"rsync $TARGET_DATE/ → NAS" "$COPIED_FILES" "$SRC_FILES" | tee -a "$LOG_FILE"
fi
else
printf " %-36s %s\n" "rsync $TARGET_DATE/ → NAS" "skipped (no source dir)"
fi
# ─── SUMMARY ──────────────────────────────────────────────────────────────────
TOTAL_ELAPSED=$(( $(date +%s) - SCRIPT_START ))
DB_SIZE=$(du -sh "$NAS_DB" | cut -f1)
echo ""
log "Archive DB size: $DB_SIZE"
log "Total time: $(( TOTAL_ELAPSED / 60 ))m$(( TOTAL_ELAPSED % 60 ))s"
log "Sync complete for $TARGET_DATE"
log "========================================"
#!/bin/bash
# screenpipe_sync.sh
# Syncs Screenpipe SQLite data to a NAS archive database (append-only, no deletions).
# Also copies the day's video/frame data folder to the NAS.
#
# Usage:
# ./screenpipe_sync.sh # syncs yesterday (default)
# ./screenpipe_sync.sh 2026-04-15 # syncs a specific date
# ./screenpipe_sync.sh today # syncs today so far
#
# Cron example (runs at 3am daily):
# 0 3 * * * /Users/lukas/.screenpipe/screenpipe_sync.sh >> /Users/lukas/.screenpipe/sync.log 2>&1
set -euo pipefail
# ─── CONFIG ───────────────────────────────────────────────────────────────────
DB_SRC="${SCREENPIPE_DB:-$HOME/.screenpipe/db.sqlite}"
NAS_MOUNT="${NAS_MOUNT:-/Volumes/Test/screenpipe}"
NAS_DB="$NAS_MOUNT/archive.db"
NAS_DATA="$NAS_MOUNT/data"
LOG_FILE="$HOME/.screenpipe/sync.log"
# ──────────────────────────────────────────────────────────────────────────────
# ─── HELPERS ──────────────────────────────────────────────────────────────────
SCRIPT_START=$(date +%s)
log() {
local msg="[$(date '+%Y-%m-%d %H:%M:%S')] $*"
echo "$msg" | tee -a "$LOG_FILE"
}
step() {
local now=$(date +%s)
local elapsed=$(( now - SCRIPT_START ))
local min=$(( elapsed / 60 ))
local sec=$(( elapsed % 60 ))
printf "\n[+%02dm%02ds] ▶ %s\n" "$min" "$sec" "$*" | tee -a "$LOG_FILE"
}
run_sqlite_heredoc() {
local label="$1"
local sql="$2"
local start=$(date +%s)
printf " %-36s " "$label"
sqlite3 "$DB_SRC" <<< "$sql" &
local pid=$!
local spin=[PASSWORD] '⠙' '⠹' '⠸' '⠼' '⠴' '⠦' '⠧' '⠇' '⠏')
local i=0
while kill -0 "$pid" 2>/dev/null; do
printf "\r %-36s %s " "$label" "${spin[$i]}"
i=$(( (i + 1) % 10 ))
sleep 0.2
done
wait "$pid"
local rc=$?
if [ $rc -ne 0 ]; then
printf "\r %-36s ✗ FAILED\n" "$label" | tee -a "$LOG_FILE"
exit $rc
fi
local dur=$(( $(date +%s) - start ))
printf "\r %-36s ✓ %dm%02ds\n" "$label" "$(( dur / 60 ))" "$(( dur % 60 ))" | tee -a "$LOG_FILE"
}
check() {
local label="$1" got="$2" expected="$3"
if [ "$got" -eq "$expected" ]; then
printf " %-20s %s / %s ✓\n" "$label:" "$got" "$expected"
else
printf " %-20s %s / %s ✗ MISMATCH\n" "$label:" "$got" "$expected"
fi
}
# ──────────────────────────────────────────────────────────────────────────────
# ─── DATE ARGUMENT ────────────────────────────────────────────────────────────
if [ "${1:-}" = "today" ]; then
TARGET_DATE=$(date +%Y-%m-%d)
elif [ -n "${1:-}" ]; then
TARGET_DATE="$1"
if ! [[ "$TARGET_DATE" =~ ^[0-9]{4}-[0-9]{2}-[0-9]{2}$ ]]; then
echo "ERROR: Invalid date format. Use YYYY-MM-DD, 'today', or no argument for yesterday."
exit 1
fi
else
TARGET_DATE=$(date -v-1d +%Y-%m-%d)
fi
log "========================================"
log "Screenpipe sync starting for: $TARGET_DATE"
log "========================================"
# ─── PREFLIGHT ────────────────────────────────────────────────────────────────
step "Preflight checks"
if [ ! -f "$DB_SRC" ]; then
log "ERROR: Source DB not found at $DB_SRC"; exit 1
fi
printf " %-20s %s (%s)\n" "Source DB:" "OK" "$(du -sh "$DB_SRC" | cut -f1)"
if [ ! -d "$NAS_MOUNT" ]; then
log "ERROR: NAS not mounted at $NAS_MOUNT"; exit 1
fi
printf " %-20s %s\n" "NAS mount:" "OK $NAS_MOUNT"
# Check if DB already synced for this date
DB_ALREADY_SYNCED=false
if [ -f "$NAS_DB" ]; then
EXISTING=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';" 2>/dev/null || echo "0")
if [ "$EXISTING" -gt "0" ]; then
log "Date $TARGET_DATE already has $EXISTING frames in archive — skipping DB sync"
DB_ALREADY_SYNCED=true
else
printf " %-20s %s (%s)\n" "Archive DB:" "exists" "$(du -sh "$NAS_DB" | cut -f1)"
fi
else
printf " %-20s %s\n" "Archive DB:" "will be created"
fi
# Source data dir for this date
DATA_SRC="$HOME/.screenpipe/data/data/$TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
DATA_SIZE=$(du -sh "$DATA_SRC" | cut -f1)
DATA_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
printf " %-20s %s (%s files, %s)\n" "Data dir:" "OK" "$DATA_FILES" "$DATA_SIZE"
else
printf " %-20s %s\n" "Data dir:" "not found — skipping file copy"
fi
# ─── DB SYNC ──────────────────────────────────────────────────────────────────
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── COUNT SOURCE ROWS ────────────────────────────────────────────────────
step "Counting source rows for $TARGET_DATE"
SRC_FRAMES=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
SRC_ELEMENTS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_UI=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
SRC_OCR=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_MEETINGS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start) = '$TARGET_DATE';")
printf " %-20s %s\n" "frames:" "$SRC_FRAMES"
printf " %-20s %s\n" "elements:" "$SRC_ELEMENTS"
printf " %-20s %s\n" "ui_events:" "$SRC_UI"
printf " %-20s %s\n" "ocr_text:" "$SRC_OCR"
printf " %-20s %s\n" "meetings:" "$SRC_MEETINGS"
if [ "$SRC_FRAMES" -eq "0" ]; then
log "No frames found for $TARGET_DATE — skipping DB sync"
DB_ALREADY_SYNCED=true
fi
fi
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── INIT TABLES ──────────────────────────────────────────────────────────
step "Initialising tables, indexes, FTS"
run_sqlite_heredoc "creating tables" "
ATTACH '$NAS_DB' AS nas;
CREATE TABLE IF NOT EXISTS nas.frames AS SELECT * FROM main.frames WHERE 0;
CREATE TABLE IF NOT EXISTS nas.elements AS SELECT * FROM main.elements WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ui_events AS SELECT * FROM main.ui_events WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ocr_text AS SELECT * FROM main.ocr_text WHERE 0;
CREATE TABLE IF NOT EXISTS nas.video_chunks AS SELECT * FROM main.video_chunks WHERE 0;
CREATE TABLE IF NOT EXISTS nas.meetings AS SELECT * FROM main.meetings WHERE 0;
DETACH nas;
"
run_sqlite_heredoc "creating indexes" "
ATTACH '$NAS_DB' AS nas;
CREATE INDEX IF NOT EXISTS nas.idx_frames_timestamp ON frames(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_frames_app_name ON frames(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_window_name ON frames(window_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_video_chunk_id ON frames(video_chunk_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_id ON elements(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_src_role ON elements(frame_id, source, role) WHERE text IS NOT NULL;
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_timestamp ON ui_events(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_app_name ON ui_events(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_frame_id ON ui_events(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_ocr_text_frame_id ON ocr_text(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_meetings_start ON meetings(meeting_start);
CREATE INDEX IF NOT EXISTS nas.idx_video_chunks_device ON video_chunks(device_name);
DETACH nas;
"
run_sqlite_heredoc "creating FTS tables" "
ATTACH '$NAS_DB' AS nas;
CREATE VIRTUAL TABLE IF NOT EXISTS nas.elements_fts USING fts5(
text, role, frame_id UNINDEXED,
content='elements', content_rowid='id', tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.frames_fts USING fts5(
full_text, app_name, window_name, browser_url, id UNINDEXED,
tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.ui_events_fts USING fts5(
text_content,
app_name,
window_title,
element_name,
content='ui_events',
content_rowid='id',
tokenize='unicode61'
);
DETACH nas;
"
# ─── SYNC DATA ────────────────────────────────────────────────────────────
step "Syncing data for $TARGET_DATE"
run_sqlite_heredoc "video_chunks" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.video_chunks
SELECT * FROM main.video_chunks
WHERE id IN (
SELECT DISTINCT video_chunk_id FROM main.frames
WHERE date(timestamp) = '$TARGET_DATE' AND video_chunk_id IS NOT NULL
);
DETACH nas;
"
run_sqlite_heredoc "frames ($SRC_FRAMES rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.frames
SELECT * FROM main.frames WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ocr_text ($SRC_OCR rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ocr_text
SELECT o.* FROM main.ocr_text o
JOIN main.frames f ON o.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ui_events ($SRC_UI rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ui_events
SELECT * FROM main.ui_events WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "elements ($SRC_ELEMENTS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.elements
SELECT e.* FROM main.elements e
JOIN main.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "meetings ($SRC_MEETINGS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.meetings
SELECT * FROM main.meetings WHERE date(meeting_start) = '$TARGET_DATE';
DETACH nas;
"
# ─── FTS UPDATE ───────────────────────────────────────────────────────────
step "Updating FTS indexes"
run_sqlite_heredoc "elements_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.elements_fts(rowid, text, role)
SELECT e.id, e.text, e.role
FROM nas.elements e
JOIN nas.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE'
AND e.text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "frames_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.frames_fts(rowid, full_text, app_name, window_name, browser_url, id)
SELECT id, full_text, app_name, window_name, browser_url, id
FROM nas.frames
WHERE date(timestamp) = '$TARGET_DATE'
AND full_text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "ui_events_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.ui_events_fts(rowid, text_content, app_name, window_title, element_name)
SELECT id, text_content, app_name, window_title, element_name
FROM nas.ui_events
WHERE date(timestamp) = '$TARGET_DATE'
AND text_content IS NOT NULL;
DETACH nas;
"
# ─── VERIFY DB ────────────────────────────────────────────────────────────
step "Verifying DB"
V_FRAMES=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
V_ELEMENTS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_UI=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
V_OCR=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_MEETINGS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start)= '$TARGET_DATE';")
check "frames" "$V_FRAMES" "$SRC_FRAMES"
check "elements" "$V_ELEMENTS" "$SRC_ELEMENTS"
check "ui_events" "$V_UI" "$SRC_UI"
check "ocr_text" "$V_OCR" "$SRC_OCR"
check "meetings" "$V_MEETINGS" "$SRC_MEETINGS"
fi
# ─── COPY DATA FOLDER ─────────────────────────────────────────────────────────
# Always runs regardless of DB sync status
step "Copying data folder for $TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
mkdir -p "$NAS_DATA/$TARGET_DATE"
RSYNC_START=$(date +%s)
printf " %-36s " "rsync $TARGET_DATE/ → NAS"
rsync -a --ignore-existing \
"$DATA_SRC/" \
"$NAS_DATA/$TARGET_DATE/" \
2>>"$LOG_FILE"
RSYNC_DUR=$(( $(date +%s) - RSYNC_START ))
COPIED_FILES=$(ls "$NAS_DATA/$TARGET_DATE" | wc -l | tr -d ' ')
SRC_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
COPIED_SIZE=$(du -sh "$NAS_DATA/$TARGET_DATE" | cut -f1)
if [ "$COPIED_FILES" -eq "$SRC_FILES" ]; then
printf "\r %-36s ✓ %dm%02ds (%s files, %s)\n" \
"rsync $TARGET_DATE/ → NAS" \
"$(( RSYNC_DUR / 60 ))" "$(( RSYNC_DUR % 60 ))" \
"$COPIED_FILES" "$COPIED_SIZE" | tee -a "$LOG_FILE"
else
printf "\r %-36s ✗ %s / %s files\n" \
"rsync $TARGET_DATE/ → NAS" "$COPIED_FILES" "$SRC_FILES" | tee -a "$LOG_FILE"
fi
else
printf " %-36s %s\n" "rsync $TARGET_DATE/ → NAS" "skipped (no source dir)"
fi
# ─── SUMMARY ──────────────────────────────────────────────────────────────────
TOTAL_ELAPSED=$(( $(date +%s) - SCRIPT_START ))
DB_SIZE=$(du -sh "$NAS_DB" | cut -f1)
echo ""
log "Archive DB size: $DB_SIZE"
log "Total time: $(( TOTAL_ELAPSED / 60 ))m$(( TOTAL_ELAPSED % 60 ))s"
log "Sync complete for $TARGET_DATE"
log "========================================"
lets add ability to jump…, Editor Group 2
Problems (⇧⌘M)
PROBLEMS
Output (⇧⌘U)
OUTPUT
Debug Console (⇧⌘Y)
DEBUG CONSOLE
Terminal (⌃`)
TERMINAL
Ports
PORTS
remote SSH: nas
SSH: nas
screenpipe (Git) - master*, Checkout Branch/Tag...
master*
screenpipe (Git) - Synchronize Changes
No Problems
0
0
No Ports Forwarded
0
Notifications
Signed out
Signed out
Shell Script
Editor Language Status: $(copilot) No inline suggestion available, Inline suggestions
LF
UTF-8
Spaces: 2
Ln 352, Col 47 (13096 selected)
Screen Reader Optimized
git-commit Lukas Kovalik (23 hours ago)
Lukas Kovalik (23 hours ago)
Info: Downloading VS Code Server...
lets add ability to jump fro,m search to timeline so we can see the exact screen (also based on monitor and time
Session history
New session
Message actions
This session is being continued from a previous conversation that ran out of context. The summary below covers the earlier portion of the conversation.
Summary:
1. Primary Request and Intent:
The user has been developing a Screenpipe dashboard on a Synology NAS. Requests in chronological order:
- **Q1**: Are db.sqlite-shm and db.sqlite-wal used? → Explained WAL mode; cleaned up files.
- **Q2**: AND combination search (term1 + term2 must both appear) → Added "AND also" input field.
- **Q3**: Explain UTC time issue, show date in results, make Only/Skip apps use multiselect → Implemented all three.
- **Q4**: Point sqlite-web at archive.db, add second sqlite-web for db.sqlite → Updated docker-compose.yml, added two services on ports 8767/8768.
- **Q5**: "still the same" (container not picking up config) → Explained `up -d` vs `restart`.
- **Q6**: Add git, push to gitea.com/lakylak/screenpipe.git → Initialized repo, committed, pushed.
- **Q7**: Add Timetable tab with video playback and timeline → Full implementation with click-to-seek, auto-advancing chunks, legend.
- **Q8**: Add scroll/zoom on timeline, jump to specific time, better awareness → Implemented zoom (1×–32×), scrollable timeline, overview minimap, viewport indicator, Jump-to-time input, screenshot thumbnail.
- **Q9 (most recent)**: Make playing speed more than 8×, fix speed not working, show position on timeline → CSS fix done (wider playhead + time label style), remainder in progress.
2. Key Technical Concepts:
- **SQLite WAL mode**: db.sqlite-shm and db.sqlite-wal are WAL companion files; cleaned with `PRAGMA wal_checkpoint(TRUNCATE)`.
- **SQLite FTS5**: Space-separated terms = AND; `MATCH 'term1 term2'` requires both; OR requires explicit `OR` keyword.
- **FastAPI**: Python backend with route handlers in main.py, consumers pattern for DB queries.
- **Docker Compose**: App image is built (code baked in), not volume-mounted → requires `build` + `up -d` for code changes. `restart` doesn't re-read compose file — need `up -d`.
- **sqlite-web**: Read-only flag `-r`, pre-built image, no rebuild needed for config changes.
- **Video playback**: screenpipe records at 0.1–0.3 fps (timelapse). `seek = offset_index / fps` gives seconds into video chunk. ~245 chunks/day across 2 monitors. Videos may be encoded at standard fps making them very short at native speed.
- **Timeline rendering**: App-colored segments built by merging consecutive frames with same `app_name`. Percentage positions: `((ts - dayStart) / span * 100)%`.
- **Zoom/scroll**: Inner div width = `zoom × 100%`, outer wrapper has `overflow-x:auto`. Playhead position stays as % of inner div, which is correct at all zoom levels.
- **Timer-based speed**: For speed > 2×, browser `playbackRate` is unreliable on low-FPS video; correct approach is `setInterval` advancing `video.currentTime` by `speed * interval_ms / 1000` seconds per tick.
- **UTC vs local time**: Backend treats `time_from`/`time_to` as UTC. Frontend converts with `localTimeToUtc()` before sending.
- **Git**: `#recycle/` in .gitignore treated as comment — must escape as `\#recycle/`.
3. Files and Code Sections:
- **`/volume1/screenpipe/app/main.py`**
- Added `/api/apps` endpoint (all unique app names from frames table)
- Added `/api/timetable` endpoint (frames + video URLs + screenshot URLs per device per date)
- `/api/timetable` response structure: `{frames: [{ts, app, win, video_url, seek, screenshot_url}], devices: [...], device: "..."}`
- Key snippet:
```python
@app.get("/api/timetable"
)
async def timetable_data(d: str | None = Query(None), device: str | None = Query(None)):
...
rows = conn.execute(f"""
SELECT f.timestamp AS ts, f.app_name AS app, f.window_name AS win,
f.offset_index, f.snapshot_path,
vc.file_path AS video_path, vc.fps AS video_fps
FROM frames f
JOIN video_chunks vc ON f.video_chunk_id = vc.id
WHERE f.timestamp >= ? AND f.timestamp < ?
AND f.video_chunk_id IS NOT NULL {dev_clause}
ORDER BY f.timestamp
""", params).fetchall()
# seek = offset_index / fps, screenshot_url from _snapshot_to_url(snapshot_path)
```
- **`/volume1/screenpipe/app/docker-compose.yml`**
- Changed sqlite-web service name to `sqlite-web-archive` → port 8767 → `/data/archive.db`
- Added `sqlite-web-live` → port 8768 → `/data/db.sqlite`
- screenpipe-app: port 8766, `DB_PATH: /data/db.sqlite`, `TZ: Europe/Sofia`
- **`/volume1/screenpipe/app/static/index.html`** (single large file with all CSS/HTML/JS)
- **Search enhancements**:
- Added `id="sq2"` "AND also" input between search-row1 and search-row2
- Replaced `sOnlyApps`/`sSkipApps` text inputs with custom multiselect components (`id="msOnlyApps"`, `id="msSkipApps"`)
- Updated `buildUrl()` to combine `q + " " + q2` and convert times with `localTimeToUtc()`
- Added `fmtDateTime(ts)` showing "Apr 20 14:08" format
- Added `localTimeToUtc(timeStr)` converting local HH:MM to UTC
- Multiselect state: `const _msState = { msOnlyApps: new Set(), msSkipApps: new Set() }`
- **Timetable tab** (new page):
- Tab button added between Work Report and AI Summary
- `showPage` array updated: `['activity','search','audio','report','timetable','summary']`
- `refresh()` calls `loadTimetable()` when on timetable page
- **Timetable CSS** (key styles):
```css
.tt-scroll-wrap { overflow-x:auto; ... scrollbar-width:thin; }
.tt-inner { position:relative; height:52px; } /* width set dynamically */
.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
.tt-overview-wrap { position:relative; height:14px; ... cursor:pointer; }
.tt-overview-vp { position:absolute; background:rgba(255,255,255,0.12); border:1px solid rgba(255,255,255,0.4); }
```
- **Timetable JS state variables**:
```javascript
let _ttFrames = [], _ttChunks = [], _ttCurChunk = -1;
let _ttDayStart = 0, _ttDayEnd = 0, _ttZoom = 1;
const _ttAppColors = {};
const TT_ZOOMS = [1, 2, 4, 8, 16, 32];
const TT_TICK_MIN = {1:30, 2:15, 4:10, 8:5, 16:2, 32:1};
```
- **Key JS functions**:
- `loadTimetable()` — fetches `/api/timetable`, groups chunks, renders
- `ttZoom(dir)` — changes zoom level, updates inner width, re-renders ticks, scrolls to playhead
- `ttTimelineClick(e)` — uses `inner.getBoundingClientRect()` (works correctly when scrolled)
- `ttOverviewClick(e)` — scrolls timeline + seeks video
- `ttUpdateOverviewVp()` — updates white viewport rect using `wrap.scrollLeft / inner.scrollWidth`
- `ttJumpToTime()` — parses time input, converts to ms, calls `ttSeekToMs`
- `ttSeekToMs(tsMs)` — binary-searches nearest frame, loads chunk, seeks video
- `_ttScrollToPlayhead(force)` — scrolls wrap so playhead stays at ~35% from left when Follow is checked
- `_ttMovePlayhead(tsMs)` — sets `ph.style.left` as percentage, calls `_ttScrollToPlayhead`
- `_ttShowNow(f)` — updates screenshot thumbnail + now-text div
- `ttSyncBtn()` — syncs play/pause button text
- Keyboard: Space=play/pause, ←/→=±10s, Ctrl+←/→=±30s, +/-=zoom
- **`/volume1/screenpipe/.gitignore`**
```
*.db, *.sqlite, *.sqlite-shm, *.sqlite-wal
data/, \#recycle/,
@eaDir/
app_settings.json, app/.env
__pycache__/, *.pyc, .DS_Store
```
4. Errors and Fixes:
- **`#recycle/` in .gitignore treated as comment**: `#` starts a comment in .gitignore. Fixed by escaping: `\#recycle/`. Also had to `git rm --cached "#recycle/desktop.ini"` to unstage it.
- **`sudo docker compose restart sqlite-web` didn't apply config change**: `restart` reuses existing container config, doesn't re-read compose file. Fix: `sudo docker compose up -d` recreates containers with new config.
- **Orphan container holding port 8767**: Old `screenpipe-sqlweb` container still running after rename. Fix: `sudo docker compose up -d --remove-orphans`.
- **`sudo docker` not accessible from Claude Code**: Interactive sudo required. User runs commands in NAS terminal.
- **`git config --global --add safe.directory /volume1/screenpipe`**: Required due to dubious ownership of NAS volume directory.
5. Problem Solving:
- **Time zone confusion in search**: Backend uses UTC for time_from/time_to; frontend displays local time. Solved by `localTimeToUtc()` converting before API call, and showing date+time in results.
- **sqlite-web not showing April 17–21 data**: Container was pointing at db.sqlite (old config), not archive.db. Fixed by updating docker-compose.yml + `up -d`.
- **Video playback speed**: Browser `video.playbackRate` is capped (~16× max in Chrome) and performs poorly for very low FPS (0.1–0.3 fps) screenpipe videos. Each chunk may only be 1–2 seconds of wall-clock encoded video, playing through instantly. Solution: implement timer-based frame-jumping for speed > 2× (in progress).
- **Playhead visibility**: White 2px line was hard to see. Fixed CSS to 3px width + stronger box-shadow + added `.tt-playhead-time` overlay label style.
6. All User Messages:
- "do we use db.sqlite-shm and db.sqlite-wal for anything right now? [image of search UI] Lets add ability to search with combination... Explain how does the search work time seems to work in utc maybe display date next to the time in results. Make only apps and skip apps multiselect with all available apps."
- "doi I need to restart?"
- "there is sudo docker compose"
- "no still the same"
- "add git. I have created gitea repo https://gitea.com/lakylak/screenpipe.git"
- "just do it" (re: git push)
- "how to make sure data in :8767 are accurate and point to archive.db"
- "I can't see older than 16th there should 21st of april. In http://[IP_ADDRESS]:8766/ I can see all but not on sqlite viewer http://[IP_ADDRESS]:8767"
- "ok lets add additional port to read db.sqlite, I would like to see the structure of official db that reads db.sqlite db and one for archive"
- [docker compose output showing orphan container error]
- "looks great. I want to have ability to scroll left and right and also to jump to specific time. I want to be more aware to what I see. Still it should be fast"
- "lets make the playing speed more than 8x"
- "still playing speed is not showing properly still just in one second. also I want to see in timeline where I am"
7. Pending Tasks:
- **Playback speed fix** (in progress, partially done):
- Add `<div class="tt-playhead-time" id="ttPlayheadTime"></div>` inside `ttInner` HTML
- Implement `_ttStartSpeedTimer(speed)` / `_ttStopSpeedTimer()` / `ttSetSpeed(speed)` functions
- Update speed select `onchange` to call `ttSetSpeed(+this.value)`
- Add 64× option to speed select
- Update `_ttMovePlayhead` to also update `ttPlayheadTime` label
- Update `ttSyncBtn` to start/stop timer based on play state + speed
- Update `ttNextChunk` and `ttSeekToMs` to stop timer (restarted by onplay → ttSyncBtn)
- Rebuild Docker image after changes
8. Current Work:
Working on fixing two issues with the Timetable tab:
1. **Playback speed**: `video.playbackRate` doesn't work well for low-FPS screenpipe videos. The plan is to implement a `setInterval`-based frame-jumper for speed > 2× that advances `video.currentTime` by `speed * intervalMs / 1000` every 100–200ms, bypassing browser playbackRate limits.
2. **Timeline position indicator**: Playhead not prominent enough.
**CSS already changed** in `/volume1/screenpipe/app/static/index.html`:
```css
.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
```
Session was interrupted by summary request after this CSS edit, mid-implementation.
9. Optional Next Step:
Continuing the fix for: "still playing speed is not showing properly still just in one second. also I want to see in timeline where I am"
The next steps in order are:
1. Add `<div class="tt-playhead-time" id="ttPlayheadTime"></div>` to the HTML inside `#ttInner` (after the existing `.tt-playhead` div)
2. Update `_ttMovePlayhead` to set `ttPlayheadTime` position and text content with `fmtTime()`
3. Add timer-based speed system to JS (new functions `_ttStartSpeedTimer`, `_ttStopSpeedTimer`, `ttSetSpeed`)
4. Update speed select's `onchange` to `ttSetSpeed(+this.value)` and add 64× option
5. Update `ttSyncBtn`, `ttNextChunk`, `ttSeekToMs` to integrate with timer
6. Rebuild: `sudo docker compose build screenpipe-app && sudo docker compose up -d screenpipe-app`
If you need specific details from before compaction (like exact code snippets, error messages, or content you generated), read the full transcript at: /home/Adm1n/.claude/projects/-volume1-screenpipe/7777f46e-1e40-4051-b4bb-f43aedd59556.jsonl
Continue the conversation from where it left off without asking the user any further questions. Resume directly — do not acknowledge the summary, do not recap what was happening, do not preface with "I'll continue" or similar. Pick up the last task as if the break never happened.
Thinking
Thinking
Bash
Find timetable speed and playhead related code
IN
grep -n "tt-playhead\|ttPlayhead\|ttInner\|_ttMovePlayhead\|ttSetSpeed\|_ttStartSpeed\|_ttStopSpeed\|ttSyncBtn\|playbackRate\|speed.*select\|spdSel\|ttSpeed" /volume1/screenpipe/app/static/index.html | head -80
Copy code to clipboard
OUT
207:.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
208:.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
580: <div class="tt-inner" id="ttInner" onclick="ttTimelineClick(event)">
583: <div class="tt-playhead" id="ttPlayhead"></div>
598: onpause="ttSyncBtn()"
599: onplay="ttSyncBtn()">
607: <select id="ttSpeed" onchange="document.getElementById('ttVideo').playbackRate=+this.value">
1624: document.getElementById('ttInner').style.width = '100%';
1632: document.getElementById('ttPlayhead').style.display = 'none';
1694: document.getElementById('ttInner').style.width = (_ttZoom * 100) + '%';
1700: const inner = document.getElementById('ttInner');
1710: const inner = document.getElementById('ttInner');
1718: const inner = document.getElementById('ttInner');
1750: _ttMovePlayhead(new Date(best.ts).getTime());
1772: _ttMovePlayhead(new Date(best.ts).getTime());
1791:function _ttMovePlayhead(tsMs) {
1792: const ph = document.getElementById('ttPlayhead');
1802: const ph = document.getElementById('ttPlayhead');
1805: const inner = document.getElementById('ttInner');
1819:function ttSyncBtn() {
Read
index.html
index.html
(lines 576-625)
Read
index.html
index.html
(lines 1786-1865)
Thinking
Thinking
Bash
Find chunk loading and playhead update functions
IN
grep -n "ttUpdatePlayhead\|ttNextChunk\|ttLoadChunk\|_ttCurChunk\|_ttChunks\|_ttFrames" /volume1/screenpipe/app/static/index.html | head -50
Copy code to clipboard
OUT
596: onended="ttNextChunk()"
597: ontimeupdate="ttUpdatePlayhead()"
1569:let _ttFrames = [];
1570:let _ttChunks = [];
1571:let _ttCurChunk = -1;
1607: _ttFrames = data.frames;
1609: for (const f of _ttFrames) {
1613: _ttChunks = [...chunkMap.values()];
1614: _ttCurChunk = -1;
1616: _ttDayStart = new Date(_ttFrames[0].ts).getTime();
1617: _ttDayEnd = new Date(_ttFrames[_ttFrames.length - 1].ts).getTime();
1620: for (const f of _ttFrames) _ttColor(f.app);
1640: for (const f of _ttFrames) {
1736: if (!_ttFrames.length) return;
1737: let best = _ttFrames[0], minDiff = Infinity;
1738: for (const f of _ttFrames) {
1742: const ci = _ttChunks.findIndex(ch => ch[0].video_url === best.video_url);
1744: _ttCurChunk = ci;
1753:function ttNextChunk() {
1754: if (_ttCurChunk + 1 >= _ttChunks.length) return;
1755: _ttCurChunk++;
1757: video.src = _ttChunks[_ttCurChunk][0].video_url;
1762:function ttUpdatePlayhead() {
1764: if (_ttCurChunk < 0 || _ttCurChunk >= _ttChunks.length) return;
1765: const chunk = _ttChunks[_ttCurChunk];
Read
index.html
index.html
(lines 1754-1793)
Thinking
Thinking
Read
index.html
index.html
(lines 1701-1760)
Thinking
Thinking
Read
index.html
index.html
(lines 1561-1640)
Thinking
Thinking
Bash
Find time formatting functions
IN
grep -n "function fmtTime\|function fmtDate\|function _fmt\|localTimeToUtc\|fmtDateTime" /volume1/screenpipe/app/static/index.html | head -20
Copy code to clipboard
OUT
1088: if (tf) url += `&time_from=${localTimeToUtc(tf)}`;
1089: if (tt) url += `&time_to=${localTimeToUtc(tt)}`;
1153: let meta = `<span style="color:var(--muted)">${fmtDateTime(r.timestamp)}</span>${srcBadge}`;
1166: meta = `<span style="color:var(--muted)">${fmtDateTime(r.timestamp)}</span>${srcBadge} <span class="badge ${r.is_input_device?'mic':'sys'}">${r.is_input_device?'MIC':'SYS'}</span>`;
1307:function fmtTime(ts) {
1312:function fmtDateTime(ts) {
1320:function localTimeToUtc(timeStr) {
1778:function _fmtSec(s) {
1788: `${fmtDateTime(f.ts)} · ${f.app}${f.win ? ' / ' + f.win : ''}`;
Read
index.html
index.html
(lines 1308-1327)
Thinking
Thinking
Now I have everything I need. I'll implement the frame-by-frame timer system that replaces
video.playbackRate
with wall-clock accurate playback, plus add the playhead time label. Let me make the changes:
Edit
index.html
index.html
Added 1 line...
|
Code
|
screenpipe_sync.sh — screenpipe [SSH: nas]
|
NULL
|
|
Explorer (⇧⌘E)
Search (⇧⌘F)
Source Control (⌃⇧ Explorer (⇧⌘E)
Search (⇧⌘F)
Source Control (⌃⇧G) - 2 pending changes
2
Run and Debug (⇧⌘D)
Remote Explorer
Extensions (⇧⌘X) - 2 require update, 1 requires restart
3
Claude Code
Containers
EXPLORER
EXPLORER
Explorer Section: screenpipe [SSH: nas]
Explorer Section: screenpipe [SSH: nas]
SCREENPIPE [SSH: NAS]
#recycle
app
data
pipes
.gitignore
app_settings.json
archive.db
db.sqlite
screenpipe_sync.sh
screenpipe.db
Outline Section
OUTLINE
OUTLINE
Timeline Section
TIMELINE
TIMELINE
app_settings.json, Editor Group 1
screenpipe_sync.sh, Editor Group 1
db.sqlite-shm, preview, Editor Group 1
#!/bin/bash
# screenpipe_sync.sh
# Syncs Screenpipe SQLite data to a NAS archive database (append-only, no deletions).
# Also copies the day's video/frame data folder to the NAS.
#
# Usage:
# ./screenpipe_sync.sh # syncs yesterday (default)
# ./screenpipe_sync.sh 2026-04-15 # syncs a specific date
# ./screenpipe_sync.sh today # syncs today so far
#
# Cron example (runs at 3am daily):
# 0 3 * * * /Users/lukas/.screenpipe/screenpipe_sync.sh >> /Users/lukas/.screenpipe/sync.log 2>&1
set -euo pipefail
# ─── CONFIG ───────────────────────────────────────────────────────────────────
DB_SRC="${SCREENPIPE_DB:-$HOME/.screenpipe/db.sqlite}"
NAS_MOUNT="${NAS_MOUNT:-/Volumes/Test/screenpipe}"
NAS_DB="$NAS_MOUNT/archive.db"
NAS_DATA="$NAS_MOUNT/data"
LOG_FILE="$HOME/.screenpipe/sync.log"
# ──────────────────────────────────────────────────────────────────────────────
# ─── HELPERS ──────────────────────────────────────────────────────────────────
SCRIPT_START=$(date +%s)
log() {
local msg="[$(date '+%Y-%m-%d %H:%M:%S')] $*"
echo "$msg" | tee -a "$LOG_FILE"
}
step() {
local now=$(date +%s)
local elapsed=$(( now - SCRIPT_START ))
local min=$(( elapsed / 60 ))
local sec=$(( elapsed % 60 ))
printf "\n[+%02dm%02ds] ▶ %s\n" "$min" "$sec" "$*" | tee -a "$LOG_FILE"
}
run_sqlite_heredoc() {
local label="$1"
local sql="$2"
local start=$(date +%s)
printf " %-36s " "$label"
sqlite3 "$DB_SRC" <<< "$sql" &
local pid=$!
local spin=[PASSWORD] '⠙' '⠹' '⠸' '⠼' '⠴' '⠦' '⠧' '⠇' '⠏')
local i=0
while kill -0 "$pid" 2>/dev/null; do
printf "\r %-36s %s " "$label" "${spin[$i]}"
i=$(( (i + 1) % 10 ))
sleep 0.2
done
wait "$pid"
local rc=$?
if [ $rc -ne 0 ]; then
printf "\r %-36s ✗ FAILED\n" "$label" | tee -a "$LOG_FILE"
exit $rc
fi
local dur=$(( $(date +%s) - start ))
printf "\r %-36s ✓ %dm%02ds\n" "$label" "$(( dur / 60 ))" "$(( dur % 60 ))" | tee -a "$LOG_FILE"
}
check() {
local label="$1" got="$2" expected="$3"
if [ "$got" -eq "$expected" ]; then
printf " %-20s %s / %s ✓\n" "$label:" "$got" "$expected"
else
printf " %-20s %s / %s ✗ MISMATCH\n" "$label:" "$got" "$expected"
fi
}
# ──────────────────────────────────────────────────────────────────────────────
# ─── DATE ARGUMENT ────────────────────────────────────────────────────────────
if [ "${1:-}" = "today" ]; then
TARGET_DATE=$(date +%Y-%m-%d)
elif [ -n "${1:-}" ]; then
TARGET_DATE="$1"
if ! [[ "$TARGET_DATE" =~ ^[0-9]{4}-[0-9]{2}-[0-9]{2}$ ]]; then
echo "ERROR: Invalid date format. Use YYYY-MM-DD, 'today', or no argument for yesterday."
exit 1
fi
else
TARGET_DATE=$(date -v-1d +%Y-%m-%d)
fi
log "========================================"
log "Screenpipe sync starting for: $TARGET_DATE"
log "========================================"
# ─── PREFLIGHT ────────────────────────────────────────────────────────────────
step "Preflight checks"
if [ ! -f "$DB_SRC" ]; then
log "ERROR: Source DB not found at $DB_SRC"; exit 1
fi
printf " %-20s %s (%s)\n" "Source DB:" "OK" "$(du -sh "$DB_SRC" | cut -f1)"
if [ ! -d "$NAS_MOUNT" ]; then
log "ERROR: NAS not mounted at $NAS_MOUNT"; exit 1
fi
printf " %-20s %s\n" "NAS mount:" "OK $NAS_MOUNT"
# Check if DB already synced for this date
DB_ALREADY_SYNCED=false
if [ -f "$NAS_DB" ]; then
EXISTING=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';" 2>/dev/null || echo "0")
if [ "$EXISTING" -gt "0" ]; then
log "Date $TARGET_DATE already has $EXISTING frames in archive — skipping DB sync"
DB_ALREADY_SYNCED=true
else
printf " %-20s %s (%s)\n" "Archive DB:" "exists" "$(du -sh "$NAS_DB" | cut -f1)"
fi
else
printf " %-20s %s\n" "Archive DB:" "will be created"
fi
# Source data dir for this date
DATA_SRC="$HOME/.screenpipe/data/data/$TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
DATA_SIZE=$(du -sh "$DATA_SRC" | cut -f1)
DATA_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
printf " %-20s %s (%s files, %s)\n" "Data dir:" "OK" "$DATA_FILES" "$DATA_SIZE"
else
printf " %-20s %s\n" "Data dir:" "not found — skipping file copy"
fi
# ─── DB SYNC ──────────────────────────────────────────────────────────────────
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── COUNT SOURCE ROWS ────────────────────────────────────────────────────
step "Counting source rows for $TARGET_DATE"
SRC_FRAMES=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
SRC_ELEMENTS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_UI=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
SRC_OCR=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_MEETINGS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start) = '$TARGET_DATE';")
printf " %-20s %s\n" "frames:" "$SRC_FRAMES"
printf " %-20s %s\n" "elements:" "$SRC_ELEMENTS"
printf " %-20s %s\n" "ui_events:" "$SRC_UI"
printf " %-20s %s\n" "ocr_text:" "$SRC_OCR"
printf " %-20s %s\n" "meetings:" "$SRC_MEETINGS"
if [ "$SRC_FRAMES" -eq "0" ]; then
log "No frames found for $TARGET_DATE — skipping DB sync"
DB_ALREADY_SYNCED=true
fi
fi
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── INIT TABLES ──────────────────────────────────────────────────────────
step "Initialising tables, indexes, FTS"
run_sqlite_heredoc "creating tables" "
ATTACH '$NAS_DB' AS nas;
CREATE TABLE IF NOT EXISTS nas.frames AS SELECT * FROM main.frames WHERE 0;
CREATE TABLE IF NOT EXISTS nas.elements AS SELECT * FROM main.elements WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ui_events AS SELECT * FROM main.ui_events WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ocr_text AS SELECT * FROM main.ocr_text WHERE 0;
CREATE TABLE IF NOT EXISTS nas.video_chunks AS SELECT * FROM main.video_chunks WHERE 0;
CREATE TABLE IF NOT EXISTS nas.meetings AS SELECT * FROM main.meetings WHERE 0;
DETACH nas;
"
run_sqlite_heredoc "creating indexes" "
ATTACH '$NAS_DB' AS nas;
CREATE INDEX IF NOT EXISTS nas.idx_frames_timestamp ON frames(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_frames_app_name ON frames(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_window_name ON frames(window_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_video_chunk_id ON frames(video_chunk_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_id ON elements(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_src_role ON elements(frame_id, source, role) WHERE text IS NOT NULL;
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_timestamp ON ui_events(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_app_name ON ui_events(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_frame_id ON ui_events(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_ocr_text_frame_id ON ocr_text(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_meetings_start ON meetings(meeting_start);
CREATE INDEX IF NOT EXISTS nas.idx_video_chunks_device ON video_chunks(device_name);
DETACH nas;
"
run_sqlite_heredoc "creating FTS tables" "
ATTACH '$NAS_DB' AS nas;
CREATE VIRTUAL TABLE IF NOT EXISTS nas.elements_fts USING fts5(
text, role, frame_id UNINDEXED,
content='elements', content_rowid='id', tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.frames_fts USING fts5(
full_text, app_name, window_name, browser_url, id UNINDEXED,
tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.ui_events_fts USING fts5(
text_content,
app_name,
window_title,
element_name,
content='ui_events',
content_rowid='id',
tokenize='unicode61'
);
DETACH nas;
"
# ─── SYNC DATA ────────────────────────────────────────────────────────────
step "Syncing data for $TARGET_DATE"
run_sqlite_heredoc "video_chunks" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.video_chunks
SELECT * FROM main.video_chunks
WHERE id IN (
SELECT DISTINCT video_chunk_id FROM main.frames
WHERE date(timestamp) = '$TARGET_DATE' AND video_chunk_id IS NOT NULL
);
DETACH nas;
"
run_sqlite_heredoc "frames ($SRC_FRAMES rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.frames
SELECT * FROM main.frames WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ocr_text ($SRC_OCR rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ocr_text
SELECT o.* FROM main.ocr_text o
JOIN main.frames f ON o.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ui_events ($SRC_UI rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ui_events
SELECT * FROM main.ui_events WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "elements ($SRC_ELEMENTS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.elements
SELECT e.* FROM main.elements e
JOIN main.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "meetings ($SRC_MEETINGS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.meetings
SELECT * FROM main.meetings WHERE date(meeting_start) = '$TARGET_DATE';
DETACH nas;
"
# ─── FTS UPDATE ───────────────────────────────────────────────────────────
step "Updating FTS indexes"
run_sqlite_heredoc "elements_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.elements_fts(rowid, text, role)
SELECT e.id, e.text, e.role
FROM nas.elements e
JOIN nas.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE'
AND e.text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "frames_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.frames_fts(rowid, full_text, app_name, window_name, browser_url, id)
SELECT id, full_text, app_name, window_name, browser_url, id
FROM nas.frames
WHERE date(timestamp) = '$TARGET_DATE'
AND full_text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "ui_events_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.ui_events_fts(rowid, text_content, app_name, window_title, element_name)
SELECT id, text_content, app_name, window_title, element_name
FROM nas.ui_events
WHERE date(timestamp) = '$TARGET_DATE'
AND text_content IS NOT NULL;
DETACH nas;
"
# ─── VERIFY DB ────────────────────────────────────────────────────────────
step "Verifying DB"
V_FRAMES=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
V_ELEMENTS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_UI=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
V_OCR=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_MEETINGS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start)= '$TARGET_DATE';")
check "frames" "$V_FRAMES" "$SRC_FRAMES"
check "elements" "$V_ELEMENTS" "$SRC_ELEMENTS"
check "ui_events" "$V_UI" "$SRC_UI"
check "ocr_text" "$V_OCR" "$SRC_OCR"
check "meetings" "$V_MEETINGS" "$SRC_MEETINGS"
fi
# ─── COPY DATA FOLDER ─────────────────────────────────────────────────────────
# Always runs regardless of DB sync status
step "Copying data folder for $TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
mkdir -p "$NAS_DATA/$TARGET_DATE"
RSYNC_START=$(date +%s)
printf " %-36s " "rsync $TARGET_DATE/ → NAS"
rsync -a --ignore-existing \
"$DATA_SRC/" \
"$NAS_DATA/$TARGET_DATE/" \
2>>"$LOG_FILE"
RSYNC_DUR=$(( $(date +%s) - RSYNC_START ))
COPIED_FILES=$(ls "$NAS_DATA/$TARGET_DATE" | wc -l | tr -d ' ')
SRC_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
COPIED_SIZE=$(du -sh "$NAS_DATA/$TARGET_DATE" | cut -f1)
if [ "$COPIED_FILES" -eq "$SRC_FILES" ]; then
printf "\r %-36s ✓ %dm%02ds (%s files, %s)\n" \
"rsync $TARGET_DATE/ → NAS" \
"$(( RSYNC_DUR / 60 ))" "$(( RSYNC_DUR % 60 ))" \
"$COPIED_FILES" "$COPIED_SIZE" | tee -a "$LOG_FILE"
else
printf "\r %-36s ✗ %s / %s files\n" \
"rsync $TARGET_DATE/ → NAS" "$COPIED_FILES" "$SRC_FILES" | tee -a "$LOG_FILE"
fi
else
printf " %-36s %s\n" "rsync $TARGET_DATE/ → NAS" "skipped (no source dir)"
fi
# ─── SUMMARY ──────────────────────────────────────────────────────────────────
TOTAL_ELAPSED=$(( $(date +%s) - SCRIPT_START ))
DB_SIZE=$(du -sh "$NAS_DB" | cut -f1)
echo ""
log "Archive DB size: $DB_SIZE"
log "Total time: $(( TOTAL_ELAPSED / 60 ))m$(( TOTAL_ELAPSED % 60 ))s"
log "Sync complete for $TARGET_DATE"
log "========================================"
#!/bin/bash
# screenpipe_sync.sh
# Syncs Screenpipe SQLite data to a NAS archive database (append-only, no deletions).
# Also copies the day's video/frame data folder to the NAS.
#
# Usage:
# ./screenpipe_sync.sh # syncs yesterday (default)
# ./screenpipe_sync.sh 2026-04-15 # syncs a specific date
# ./screenpipe_sync.sh today # syncs today so far
#
# Cron example (runs at 3am daily):
# 0 3 * * * /Users/lukas/.screenpipe/screenpipe_sync.sh >> /Users/lukas/.screenpipe/sync.log 2>&1
set -euo pipefail
# ─── CONFIG ───────────────────────────────────────────────────────────────────
DB_SRC="${SCREENPIPE_DB:-$HOME/.screenpipe/db.sqlite}"
NAS_MOUNT="${NAS_MOUNT:-/Volumes/Test/screenpipe}"
NAS_DB="$NAS_MOUNT/archive.db"
NAS_DATA="$NAS_MOUNT/data"
LOG_FILE="$HOME/.screenpipe/sync.log"
# ──────────────────────────────────────────────────────────────────────────────
# ─── HELPERS ──────────────────────────────────────────────────────────────────
SCRIPT_START=$(date +%s)
log() {
local msg="[$(date '+%Y-%m-%d %H:%M:%S')] $*"
echo "$msg" | tee -a "$LOG_FILE"
}
step() {
local now=$(date +%s)
local elapsed=$(( now - SCRIPT_START ))
local min=$(( elapsed / 60 ))
local sec=$(( elapsed % 60 ))
printf "\n[+%02dm%02ds] ▶ %s\n" "$min" "$sec" "$*" | tee -a "$LOG_FILE"
}
run_sqlite_heredoc() {
local label="$1"
local sql="$2"
local start=$(date +%s)
printf " %-36s " "$label"
sqlite3 "$DB_SRC" <<< "$sql" &
local pid=$!
local spin=[PASSWORD] '⠙' '⠹' '⠸' '⠼' '⠴' '⠦' '⠧' '⠇' '⠏')
local i=0
while kill -0 "$pid" 2>/dev/null; do
printf "\r %-36s %s " "$label" "${spin[$i]}"
i=$(( (i + 1) % 10 ))
sleep 0.2
done
wait "$pid"
local rc=$?
if [ $rc -ne 0 ]; then
printf "\r %-36s ✗ FAILED\n" "$label" | tee -a "$LOG_FILE"
exit $rc
fi
local dur=$(( $(date +%s) - start ))
printf "\r %-36s ✓ %dm%02ds\n" "$label" "$(( dur / 60 ))" "$(( dur % 60 ))" | tee -a "$LOG_FILE"
}
check() {
local label="$1" got="$2" expected="$3"
if [ "$got" -eq "$expected" ]; then
printf " %-20s %s / %s ✓\n" "$label:" "$got" "$expected"
else
printf " %-20s %s / %s ✗ MISMATCH\n" "$label:" "$got" "$expected"
fi
}
# ──────────────────────────────────────────────────────────────────────────────
# ─── DATE ARGUMENT ────────────────────────────────────────────────────────────
if [ "${1:-}" = "today" ]; then
TARGET_DATE=$(date +%Y-%m-%d)
elif [ -n "${1:-}" ]; then
TARGET_DATE="$1"
if ! [[ "$TARGET_DATE" =~ ^[0-9]{4}-[0-9]{2}-[0-9]{2}$ ]]; then
echo "ERROR: Invalid date format. Use YYYY-MM-DD, 'today', or no argument for yesterday."
exit 1
fi
else
TARGET_DATE=$(date -v-1d +%Y-%m-%d)
fi
log "========================================"
log "Screenpipe sync starting for: $TARGET_DATE"
log "========================================"
# ─── PREFLIGHT ────────────────────────────────────────────────────────────────
step "Preflight checks"
if [ ! -f "$DB_SRC" ]; then
log "ERROR: Source DB not found at $DB_SRC"; exit 1
fi
printf " %-20s %s (%s)\n" "Source DB:" "OK" "$(du -sh "$DB_SRC" | cut -f1)"
if [ ! -d "$NAS_MOUNT" ]; then
log "ERROR: NAS not mounted at $NAS_MOUNT"; exit 1
fi
printf " %-20s %s\n" "NAS mount:" "OK $NAS_MOUNT"
# Check if DB already synced for this date
DB_ALREADY_SYNCED=false
if [ -f "$NAS_DB" ]; then
EXISTING=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';" 2>/dev/null || echo "0")
if [ "$EXISTING" -gt "0" ]; then
log "Date $TARGET_DATE already has $EXISTING frames in archive — skipping DB sync"
DB_ALREADY_SYNCED=true
else
printf " %-20s %s (%s)\n" "Archive DB:" "exists" "$(du -sh "$NAS_DB" | cut -f1)"
fi
else
printf " %-20s %s\n" "Archive DB:" "will be created"
fi
# Source data dir for this date
DATA_SRC="$HOME/.screenpipe/data/data/$TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
DATA_SIZE=$(du -sh "$DATA_SRC" | cut -f1)
DATA_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
printf " %-20s %s (%s files, %s)\n" "Data dir:" "OK" "$DATA_FILES" "$DATA_SIZE"
else
printf " %-20s %s\n" "Data dir:" "not found — skipping file copy"
fi
# ─── DB SYNC ──────────────────────────────────────────────────────────────────
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── COUNT SOURCE ROWS ────────────────────────────────────────────────────
step "Counting source rows for $TARGET_DATE"
SRC_FRAMES=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
SRC_ELEMENTS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_UI=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
SRC_OCR=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_MEETINGS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start) = '$TARGET_DATE';")
printf " %-20s %s\n" "frames:" "$SRC_FRAMES"
printf " %-20s %s\n" "elements:" "$SRC_ELEMENTS"
printf " %-20s %s\n" "ui_events:" "$SRC_UI"
printf " %-20s %s\n" "ocr_text:" "$SRC_OCR"
printf " %-20s %s\n" "meetings:" "$SRC_MEETINGS"
if [ "$SRC_FRAMES" -eq "0" ]; then
log "No frames found for $TARGET_DATE — skipping DB sync"
DB_ALREADY_SYNCED=true
fi
fi
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── INIT TABLES ──────────────────────────────────────────────────────────
step "Initialising tables, indexes, FTS"
run_sqlite_heredoc "creating tables" "
ATTACH '$NAS_DB' AS nas;
CREATE TABLE IF NOT EXISTS nas.frames AS SELECT * FROM main.frames WHERE 0;
CREATE TABLE IF NOT EXISTS nas.elements AS SELECT * FROM main.elements WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ui_events AS SELECT * FROM main.ui_events WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ocr_text AS SELECT * FROM main.ocr_text WHERE 0;
CREATE TABLE IF NOT EXISTS nas.video_chunks AS SELECT * FROM main.video_chunks WHERE 0;
CREATE TABLE IF NOT EXISTS nas.meetings AS SELECT * FROM main.meetings WHERE 0;
DETACH nas;
"
run_sqlite_heredoc "creating indexes" "
ATTACH '$NAS_DB' AS nas;
CREATE INDEX IF NOT EXISTS nas.idx_frames_timestamp ON frames(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_frames_app_name ON frames(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_window_name ON frames(window_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_video_chunk_id ON frames(video_chunk_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_id ON elements(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_src_role ON elements(frame_id, source, role) WHERE text IS NOT NULL;
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_timestamp ON ui_events(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_app_name ON ui_events(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_frame_id ON ui_events(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_ocr_text_frame_id ON ocr_text(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_meetings_start ON meetings(meeting_start);
CREATE INDEX IF NOT EXISTS nas.idx_video_chunks_device ON video_chunks(device_name);
DETACH nas;
"
run_sqlite_heredoc "creating FTS tables" "
ATTACH '$NAS_DB' AS nas;
CREATE VIRTUAL TABLE IF NOT EXISTS nas.elements_fts USING fts5(
text, role, frame_id UNINDEXED,
content='elements', content_rowid='id', tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.frames_fts USING fts5(
full_text, app_name, window_name, browser_url, id UNINDEXED,
tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.ui_events_fts USING fts5(
text_content,
app_name,
window_title,
element_name,
content='ui_events',
content_rowid='id',
tokenize='unicode61'
);
DETACH nas;
"
# ─── SYNC DATA ────────────────────────────────────────────────────────────
step "Syncing data for $TARGET_DATE"
run_sqlite_heredoc "video_chunks" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.video_chunks
SELECT * FROM main.video_chunks
WHERE id IN (
SELECT DISTINCT video_chunk_id FROM main.frames
WHERE date(timestamp) = '$TARGET_DATE' AND video_chunk_id IS NOT NULL
);
DETACH nas;
"
run_sqlite_heredoc "frames ($SRC_FRAMES rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.frames
SELECT * FROM main.frames WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ocr_text ($SRC_OCR rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ocr_text
SELECT o.* FROM main.ocr_text o
JOIN main.frames f ON o.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ui_events ($SRC_UI rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ui_events
SELECT * FROM main.ui_events WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "elements ($SRC_ELEMENTS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.elements
SELECT e.* FROM main.elements e
JOIN main.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "meetings ($SRC_MEETINGS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.meetings
SELECT * FROM main.meetings WHERE date(meeting_start) = '$TARGET_DATE';
DETACH nas;
"
# ─── FTS UPDATE ───────────────────────────────────────────────────────────
step "Updating FTS indexes"
run_sqlite_heredoc "elements_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.elements_fts(rowid, text, role)
SELECT e.id, e.text, e.role
FROM nas.elements e
JOIN nas.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE'
AND e.text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "frames_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.frames_fts(rowid, full_text, app_name, window_name, browser_url, id)
SELECT id, full_text, app_name, window_name, browser_url, id
FROM nas.frames
WHERE date(timestamp) = '$TARGET_DATE'
AND full_text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "ui_events_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.ui_events_fts(rowid, text_content, app_name, window_title, element_name)
SELECT id, text_content, app_name, window_title, element_name
FROM nas.ui_events
WHERE date(timestamp) = '$TARGET_DATE'
AND text_content IS NOT NULL;
DETACH nas;
"
# ─── VERIFY DB ────────────────────────────────────────────────────────────
step "Verifying DB"
V_FRAMES=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
V_ELEMENTS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_UI=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
V_OCR=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_MEETINGS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start)= '$TARGET_DATE';")
check "frames" "$V_FRAMES" "$SRC_FRAMES"
check "elements" "$V_ELEMENTS" "$SRC_ELEMENTS"
check "ui_events" "$V_UI" "$SRC_UI"
check "ocr_text" "$V_OCR" "$SRC_OCR"
check "meetings" "$V_MEETINGS" "$SRC_MEETINGS"
fi
# ─── COPY DATA FOLDER ─────────────────────────────────────────────────────────
# Always runs regardless of DB sync status
step "Copying data folder for $TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
mkdir -p "$NAS_DATA/$TARGET_DATE"
RSYNC_START=$(date +%s)
printf " %-36s " "rsync $TARGET_DATE/ → NAS"
rsync -a --ignore-existing \
"$DATA_SRC/" \
"$NAS_DATA/$TARGET_DATE/" \
2>>"$LOG_FILE"
RSYNC_DUR=$(( $(date +%s) - RSYNC_START ))
COPIED_FILES=$(ls "$NAS_DATA/$TARGET_DATE" | wc -l | tr -d ' ')
SRC_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
COPIED_SIZE=$(du -sh "$NAS_DATA/$TARGET_DATE" | cut -f1)
if [ "$COPIED_FILES" -eq "$SRC_FILES" ]; then
printf "\r %-36s ✓ %dm%02ds (%s files, %s)\n" \
"rsync $TARGET_DATE/ → NAS" \
"$(( RSYNC_DUR / 60 ))" "$(( RSYNC_DUR % 60 ))" \
"$COPIED_FILES" "$COPIED_SIZE" | tee -a "$LOG_FILE"
else
printf "\r %-36s ✗ %s / %s files\n" \
"rsync $TARGET_DATE/ → NAS" "$COPIED_FILES" "$SRC_FILES" | tee -a "$LOG_FILE"
fi
else
printf " %-36s %s\n" "rsync $TARGET_DATE/ → NAS" "skipped (no source dir)"
fi
# ─── SUMMARY ──────────────────────────────────────────────────────────────────
TOTAL_ELAPSED=$(( $(date +%s) - SCRIPT_START ))
DB_SIZE=$(du -sh "$NAS_DB" | cut -f1)
echo ""
log "Archive DB size: $DB_SIZE"
log "Total time: $(( TOTAL_ELAPSED / 60 ))m$(( TOTAL_ELAPSED % 60 ))s"
log "Sync complete for $TARGET_DATE"
log "========================================"
lets add ability to jump…, Editor Group 2
Problems (⇧⌘M)
PROBLEMS
Output (⇧⌘U)
OUTPUT
Debug Console (⇧⌘Y)
DEBUG CONSOLE
Terminal (⌃`)
TERMINAL
Ports
PORTS
remote SSH: nas
SSH: nas
screenpipe (Git) - master*, Checkout Branch/Tag...
master*
screenpipe (Git) - Synchronize Changes
No Problems
0
0
No Ports Forwarded
0
Notifications
Signed out
Signed out
Shell Script
Editor Language Status: $(copilot) No inline suggestion available, Inline suggestions
LF
UTF-8
Spaces: 2
Ln 320, Col 36
Screen Reader Optimized
git-commit Lukas Kovalik (23 hours ago)
Lukas Kovalik (23 hours ago)
Info: Downloading VS Code Server...
lets add ability to jump fro,m search to timeline so we can see the exact screen (also based on monitor and time
Session history
New session
Message actions
This session is being continued from a previous conversation that ran out of context. The summary below covers the earlier portion of the conversation.
Summary:
1. Primary Request and Intent:
The user has been developing a Screenpipe dashboard on a Synology NAS. Requests in chronological order:
- **Q1**: Are db.sqlite-shm and db.sqlite-wal used? → Explained WAL mode; cleaned up files.
- **Q2**: AND combination search (term1 + term2 must both appear) → Added "AND also" input field.
- **Q3**: Explain UTC time issue, show date in results, make Only/Skip apps use multiselect → Implemented all three.
- **Q4**: Point sqlite-web at archive.db, add second sqlite-web for db.sqlite → Updated docker-compose.yml, added two services on ports 8767/8768.
- **Q5**: "still the same" (container not picking up config) → Explained `up -d` vs `restart`.
- **Q6**: Add git, push to gitea.com/lakylak/screenpipe.git → Initialized repo, committed, pushed.
- **Q7**: Add Timetable tab with video playback and timeline → Full implementation with click-to-seek, auto-advancing chunks, legend.
- **Q8**: Add scroll/zoom on timeline, jump to specific time, better awareness → Implemented zoom (1×–32×), scrollable timeline, overview minimap, viewport indicator, Jump-to-time input, screenshot thumbnail.
- **Q9 (most recent)**: Make playing speed more than 8×, fix speed not working, show position on timeline → CSS fix done (wider playhead + time label style), remainder in progress.
2. Key Technical Concepts:
- **SQLite WAL mode**: db.sqlite-shm and db.sqlite-wal are WAL companion files; cleaned with `PRAGMA wal_checkpoint(TRUNCATE)`.
- **SQLite FTS5**: Space-separated terms = AND; `MATCH 'term1 term2'` requires both; OR requires explicit `OR` keyword.
- **FastAPI**: Python backend with route handlers in main.py, consumers pattern for DB queries.
- **Docker Compose**: App image is built (code baked in), not volume-mounted → requires `build` + `up -d` for code changes. `restart` doesn't re-read compose file — need `up -d`.
- **sqlite-web**: Read-only flag `-r`, pre-built image, no rebuild needed for config changes.
- **Video playback**: screenpipe records at 0.1–0.3 fps (timelapse). `seek = offset_index / fps` gives seconds into video chunk. ~245 chunks/day across 2 monitors. Videos may be encoded at standard fps making them very short at native speed.
- **Timeline rendering**: App-colored segments built by merging consecutive frames with same `app_name`. Percentage positions: `((ts - dayStart) / span * 100)%`.
- **Zoom/scroll**: Inner div width = `zoom × 100%`, outer wrapper has `overflow-x:auto`. Playhead position stays as % of inner div, which is correct at all zoom levels.
- **Timer-based speed**: For speed > 2×, browser `playbackRate` is unreliable on low-FPS video; correct approach is `setInterval` advancing `video.currentTime` by `speed * interval_ms / 1000` seconds per tick.
- **UTC vs local time**: Backend treats `time_from`/`time_to` as UTC. Frontend converts with `localTimeToUtc()` before sending.
- **Git**: `#recycle/` in .gitignore treated as comment — must escape as `\#recycle/`.
3. Files and Code Sections:
- **`/volume1/screenpipe/app/main.py`**
- Added `/api/apps` endpoint (all unique app names from frames table)
- Added `/api/timetable` endpoint (frames + video URLs + screenshot URLs per device per date)
- `/api/timetable` response structure: `{frames: [{ts, app, win, video_url, seek, screenshot_url}], devices: [...], device: "..."}`
- Key snippet:
```python
@app.get("/api/timetable"
)
async def timetable_data(d: str | None = Query(None), device: str | None = Query(None)):
...
rows = conn.execute(f"""
SELECT f.timestamp AS ts, f.app_name AS app, f.window_name AS win,
f.offset_index, f.snapshot_path,
vc.file_path AS video_path, vc.fps AS video_fps
FROM frames f
JOIN video_chunks vc ON f.video_chunk_id = vc.id
WHERE f.timestamp >= ? AND f.timestamp < ?
AND f.video_chunk_id IS NOT NULL {dev_clause}
ORDER BY f.timestamp
""", params).fetchall()
# seek = offset_index / fps, screenshot_url from _snapshot_to_url(snapshot_path)
```
- **`/volume1/screenpipe/app/docker-compose.yml`**
- Changed sqlite-web service name to `sqlite-web-archive` → port 8767 → `/data/archive.db`
- Added `sqlite-web-live` → port 8768 → `/data/db.sqlite`
- screenpipe-app: port 8766, `DB_PATH: /data/db.sqlite`, `TZ: Europe/Sofia`
- **`/volume1/screenpipe/app/static/index.html`** (single large file with all CSS/HTML/JS)
- **Search enhancements**:
- Added `id="sq2"` "AND also" input between search-row1 and search-row2
- Replaced `sOnlyApps`/`sSkipApps` text inputs with custom multiselect components (`id="msOnlyApps"`, `id="msSkipApps"`)
- Updated `buildUrl()` to combine `q + " " + q2` and convert times with `localTimeToUtc()`
- Added `fmtDateTime(ts)` showing "Apr 20 14:08" format
- Added `localTimeToUtc(timeStr)` converting local HH:MM to UTC
- Multiselect state: `const _msState = { msOnlyApps: new Set(), msSkipApps: new Set() }`
- **Timetable tab** (new page):
- Tab button added between Work Report and AI Summary
- `showPage` array updated: `['activity','search','audio','report','timetable','summary']`
- `refresh()` calls `loadTimetable()` when on timetable page
- **Timetable CSS** (key styles):
```css
.tt-scroll-wrap { overflow-x:auto; ... scrollbar-width:thin; }
.tt-inner { position:relative; height:52px; } /* width set dynamically */
.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
.tt-overview-wrap { position:relative; height:14px; ... cursor:pointer; }
.tt-overview-vp { position:absolute; background:rgba(255,255,255,0.12); border:1px solid rgba(255,255,255,0.4); }
```
- **Timetable JS state variables**:
```javascript
let _ttFrames = [], _ttChunks = [], _ttCurChunk = -1;
let _ttDayStart = 0, _ttDayEnd = 0, _ttZoom = 1;
const _ttAppColors = {};
const TT_ZOOMS = [1, 2, 4, 8, 16, 32];
const TT_TICK_MIN = {1:30, 2:15, 4:10, 8:5, 16:2, 32:1};
```
- **Key JS functions**:
- `loadTimetable()` — fetches `/api/timetable`, groups chunks, renders
- `ttZoom(dir)` — changes zoom level, updates inner width, re-renders ticks, scrolls to playhead
- `ttTimelineClick(e)` — uses `inner.getBoundingClientRect()` (works correctly when scrolled)
- `ttOverviewClick(e)` — scrolls timeline + seeks video
- `ttUpdateOverviewVp()` — updates white viewport rect using `wrap.scrollLeft / inner.scrollWidth`
- `ttJumpToTime()` — parses time input, converts to ms, calls `ttSeekToMs`
- `ttSeekToMs(tsMs)` — binary-searches nearest frame, loads chunk, seeks video
- `_ttScrollToPlayhead(force)` — scrolls wrap so playhead stays at ~35% from left when Follow is checked
- `_ttMovePlayhead(tsMs)` — sets `ph.style.left` as percentage, calls `_ttScrollToPlayhead`
- `_ttShowNow(f)` — updates screenshot thumbnail + now-text div
- `ttSyncBtn()` — syncs play/pause button text
- Keyboard: Space=play/pause, ←/→=±10s, Ctrl+←/→=±30s, +/-=zoom
- **`/volume1/screenpipe/.gitignore`**
```
*.db, *.sqlite, *.sqlite-shm, *.sqlite-wal
data/, \#recycle/,
@eaDir/
app_settings.json, app/.env
__pycache__/, *.pyc, .DS_Store
```
4. Errors and Fixes:
- **`#recycle/` in .gitignore treated as comment**: `#` starts a comment in .gitignore. Fixed by escaping: `\#recycle/`. Also had to `git rm --cached "#recycle/desktop.ini"` to unstage it.
- **`sudo docker compose restart sqlite-web` didn't apply config change**: `restart` reuses existing container config, doesn't re-read compose file. Fix: `sudo docker compose up -d` recreates containers with new config.
- **Orphan container holding port 8767**: Old `screenpipe-sqlweb` container still running after rename. Fix: `sudo docker compose up -d --remove-orphans`.
- **`sudo docker` not accessible from Claude Code**: Interactive sudo required. User runs commands in NAS terminal.
- **`git config --global --add safe.directory /volume1/screenpipe`**: Required due to dubious ownership of NAS volume directory.
5. Problem Solving:
- **Time zone confusion in search**: Backend uses UTC for time_from/time_to; frontend displays local time. Solved by `localTimeToUtc()` converting before API call, and showing date+time in results.
- **sqlite-web not showing April 17–21 data**: Container was pointing at db.sqlite (old config), not archive.db. Fixed by updating docker-compose.yml + `up -d`.
- **Video playback speed**: Browser `video.playbackRate` is capped (~16× max in Chrome) and performs poorly for very low FPS (0.1–0.3 fps) screenpipe videos. Each chunk may only be 1–2 seconds of wall-clock encoded video, playing through instantly. Solution: implement timer-based frame-jumping for speed > 2× (in progress).
- **Playhead visibility**: White 2px line was hard to see. Fixed CSS to 3px width + stronger box-shadow + added `.tt-playhead-time` overlay label style.
6. All User Messages:
- "do we use db.sqlite-shm and db.sqlite-wal for anything right now? [image of search UI] Lets add ability to search with combination... Explain how does the search work time seems to work in utc maybe display date next to the time in results. Make only apps and skip apps multiselect with all available apps."
- "doi I need to restart?"
- "there is sudo docker compose"
- "no still the same"
- "add git. I have created gitea repo https://gitea.com/lakylak/screenpipe.git"
- "just do it" (re: git push)
- "how to make sure data in :8767 are accurate and point to archive.db"
- "I can't see older than 16th there should 21st of april. In http://[IP_ADDRESS]:8766/ I can see all but not on sqlite viewer http://[IP_ADDRESS]:8767"
- "ok lets add additional port to read db.sqlite, I would like to see the structure of official db that reads db.sqlite db and one for archive"
- [docker compose output showing orphan container error]
- "looks great. I want to have ability to scroll left and right and also to jump to specific time. I want to be more aware to what I see. Still it should be fast"
- "lets make the playing speed more than 8x"
- "still playing speed is not showing properly still just in one second. also I want to see in timeline where I am"
7. Pending Tasks:
- **Playback speed fix** (in progress, partially done):
- Add `<div class="tt-playhead-time" id="ttPlayheadTime"></div>` inside `ttInner` HTML
- Implement `_ttStartSpeedTimer(speed)` / `_ttStopSpeedTimer()` / `ttSetSpeed(speed)` functions
- Update speed select `onchange` to call `ttSetSpeed(+this.value)`
- Add 64× option to speed select
- Update `_ttMovePlayhead` to also update `ttPlayheadTime` label
- Update `ttSyncBtn` to start/stop timer based on play state + speed
- Update `ttNextChunk` and `ttSeekToMs` to stop timer (restarted by onplay → ttSyncBtn)
- Rebuild Docker image after changes
8. Current Work:
Working on fixing two issues with the Timetable tab:
1. **Playback speed**: `video.playbackRate` doesn't work well for low-FPS screenpipe videos. The plan is to implement a `setInterval`-based frame-jumper for speed > 2× that advances `video.currentTime` by `speed * intervalMs / 1000` every 100–200ms, bypassing browser playbackRate limits.
2. **Timeline position indicator**: Playhead not prominent enough.
**CSS already changed** in `/volume1/screenpipe/app/static/index.html`:
```css
.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
```
Session was interrupted by summary request after this CSS edit, mid-implementation.
9. Optional Next Step:
Continuing the fix for: "still playing speed is not showing properly still just in one second. also I want to see in timeline where I am"
The next steps in order are:
1. Add `<div class="tt-playhead-time" id="ttPlayheadTime"></div>` to the HTML inside `#ttInner` (after the existing `.tt-playhead` div)
2. Update `_ttMovePlayhead` to set `ttPlayheadTime` position and text content with `fmtTime()`
3. Add timer-based speed system to JS (new functions `_ttStartSpeedTimer`, `_ttStopSpeedTimer`, `ttSetSpeed`)
4. Update speed select's `onchange` to `ttSetSpeed(+this.value)` and add 64× option
5. Update `ttSyncBtn`, `ttNextChunk`, `ttSeekToMs` to integrate with timer
6. Rebuild: `sudo docker compose build screenpipe-app && sudo docker compose up -d screenpipe-app`
If you need specific details from before compaction (like exact code snippets, error messages, or content you generated), read the full transcript at: /home/Adm1n/.claude/projects/-volume1-screenpipe/7777f46e-1e40-4051-b4bb-f43aedd59556.jsonl
Continue the conversation from where it left off without asking the user any further questions. Resume directly — do not acknowledge the summary, do not recap what was happening, do not preface with "I'll continue" or similar. Pick up the last task as if the break never happened.
Thinking
Thinking
Bash
Find timetable speed and playhead related code
IN
grep -n "tt-playhead\|ttPlayhead\|ttInner\|_ttMovePlayhead\|ttSetSpeed\|_ttStartSpeed\|_ttStopSpeed\|ttSyncBtn\|playbackRate\|speed.*select\|spdSel\|ttSpeed" /volume1/screenpipe/app/static/index.html | head -80
Copy code to clipboard
OUT
207:.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
208:.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
580: <div class="tt-inner" id="ttInner" onclick="ttTimelineClick(event)">
583: <div class="tt-playhead" id="ttPlayhead"></div>
598: onpause="ttSyncBtn()"
599: onplay="ttSyncBtn()">
607: <select id="ttSpeed" onchange="document.getElementById('ttVideo').playbackRate=+this.value">
1624: document.getElementById('ttInner').style.width = '100%';
1632: document.getElementById('ttPlayhead').style.display = 'none';
1694: document.getElementById('ttInner').style.width = (_ttZoom * 100) + '%';
1700: const inner = document.getElementById('ttInner');
1710: const inner = document.getElementById('ttInner');
1718: const inner = document.getElementById('ttInner');
1750: _ttMovePlayhead(new Date(best.ts).getTime());
1772: _ttMovePlayhead(new Date(best.ts).getTime());
1791:function _ttMovePlayhead(tsMs) {
1792: const ph = document.getElementById('ttPlayhead');
1802: const ph = document.getElementById('ttPlayhead');
1805: const inner = document.getElementById('ttInner');
1819:function ttSyncBtn() {
Read
index.html
index.html
(lines 576-625)
Read
index.html
index.html
(lines 1786-1865)
Thinking
Thinking
Bash
Find chunk loading and playhead update functions
IN
grep -n "ttUpdatePlayhead\|ttNextChunk\|ttLoadChunk\|_ttCurChunk\|_ttChunks\|_ttFrames" /volume1/screenpipe/app/static/index.html | head -50
Copy code to clipboard
OUT
596: onended="ttNextChunk()"
597: ontimeupdate="ttUpdatePlayhead()"
1569:let _ttFrames = [];
1570:let _ttChunks = [];
1571:let _ttCurChunk = -1;
1607: _ttFrames = data.frames;
1609: for (const f of _ttFrames) {
1613: _ttChunks = [...chunkMap.values()];
1614: _ttCurChunk = -1;
1616: _ttDayStart = new Date(_ttFrames[0].ts).getTime();
1617: _ttDayEnd = new Date(_ttFrames[_ttFrames.length - 1].ts).getTime();
1620: for (const f of _ttFrames) _ttColor(f.app);
1640: for (const f of _ttFrames) {
1736: if (!_ttFrames.length) return;
1737: let best = _ttFrames[0], minDiff = Infinity;
1738: for (const f of _ttFrames) {
1742: const ci = _ttChunks.findIndex(ch => ch[0].video_url === best.video_url);
1744: _ttCurChunk = ci;
1753:function ttNextChunk() {
1754: if (_ttCurChunk + 1 >= _ttChunks.length) return;
1755: _ttCurChunk++;
1757: video.src = _ttChunks[_ttCurChunk][0].video_url;
1762:function ttUpdatePlayhead() {
1764: if (_ttCurChunk < 0 || _ttCurChunk >= _ttChunks.length) return;
1765: const chunk = _ttChunks[_ttCurChunk];
Read
index.html
index.html
(lines 1754-1793)
Thinking
Thinking
Read
index.html
index.html
(lines 1701-1760)
Thinking
Thinking
Read
index.html
index.html
(lines 1561-1640)
Thinking
Thinking
Bash
Find time formatting functions
IN
grep -n "function fmtTime\|function fmtDate\|function _fmt\|localTimeToUtc\|fmtDateTime" /volume1/screenpipe/app/static/index.html | head -20
Copy code to clipboard
OUT...
|
Code
|
screenpipe_sync.sh — screenpipe [SSH: nas]
|
NULL
|
|
Explorer (⇧⌘E)
Search (⇧⌘F)
Source Control (⌃⇧ Explorer (⇧⌘E)
Search (⇧⌘F)
Source Control (⌃⇧G) - 2 pending changes
2
Run and Debug (⇧⌘D)
Remote Explorer
Extensions (⇧⌘X) - 2 require update, 1 requires restart
3
Claude Code
Containers
EXPLORER
EXPLORER
Explorer Section: screenpipe [SSH: nas]
Explorer Section: screenpipe [SSH: nas]
SCREENPIPE [SSH: NAS]
#recycle
app
data
pipes
.gitignore
app_settings.json
archive.db
db.sqlite
screenpipe_sync.sh
screenpipe.db
Outline Section
OUTLINE
OUTLINE
Timeline Section
TIMELINE
TIMELINE
app_settings.json, Editor Group 1
screenpipe_sync.sh, Editor Group 1
db.sqlite-shm, preview, Editor Group 1
#!/bin/bash
# screenpipe_sync.sh
# Syncs Screenpipe SQLite data to a NAS archive database (append-only, no deletions).
# Also copies the day's video/frame data folder to the NAS.
#
# Usage:
# ./screenpipe_sync.sh # syncs yesterday (default)
# ./screenpipe_sync.sh 2026-04-15 # syncs a specific date
# ./screenpipe_sync.sh today # syncs today so far
#
# Cron example (runs at 3am daily):
# 0 3 * * * /Users/lukas/.screenpipe/screenpipe_sync.sh >> /Users/lukas/.screenpipe/sync.log 2>&1
set -euo pipefail
# ─── CONFIG ───────────────────────────────────────────────────────────────────
DB_SRC="${SCREENPIPE_DB:-$HOME/.screenpipe/db.sqlite}"
NAS_MOUNT="${NAS_MOUNT:-/Volumes/Test/screenpipe}"
NAS_DB="$NAS_MOUNT/archive.db"
NAS_DATA="$NAS_MOUNT/data"
LOG_FILE="$HOME/.screenpipe/sync.log"
# ──────────────────────────────────────────────────────────────────────────────
# ─── HELPERS ──────────────────────────────────────────────────────────────────
SCRIPT_START=$(date +%s)
log() {
local msg="[$(date '+%Y-%m-%d %H:%M:%S')] $*"
echo "$msg" | tee -a "$LOG_FILE"
}
step() {
local now=$(date +%s)
local elapsed=$(( now - SCRIPT_START ))
local min=$(( elapsed / 60 ))
local sec=$(( elapsed % 60 ))
printf "\n[+%02dm%02ds] ▶ %s\n" "$min" "$sec" "$*" | tee -a "$LOG_FILE"
}
run_sqlite_heredoc() {
local label="$1"
local sql="$2"
local start=$(date +%s)
printf " %-36s " "$label"
sqlite3 "$DB_SRC" <<< "$sql" &
local pid=$!
local spin=[PASSWORD] '⠙' '⠹' '⠸' '⠼' '⠴' '⠦' '⠧' '⠇' '⠏')
local i=0
while kill -0 "$pid" 2>/dev/null; do
printf "\r %-36s %s " "$label" "${spin[$i]}"
i=$(( (i + 1) % 10 ))
sleep 0.2
done
wait "$pid"
local rc=$?
if [ $rc -ne 0 ]; then
printf "\r %-36s ✗ FAILED\n" "$label" | tee -a "$LOG_FILE"
exit $rc
fi
local dur=$(( $(date +%s) - start ))
printf "\r %-36s ✓ %dm%02ds\n" "$label" "$(( dur / 60 ))" "$(( dur % 60 ))" | tee -a "$LOG_FILE"
}
check() {
local label="$1" got="$2" expected="$3"
if [ "$got" -eq "$expected" ]; then
printf " %-20s %s / %s ✓\n" "$label:" "$got" "$expected"
else
printf " %-20s %s / %s ✗ MISMATCH\n" "$label:" "$got" "$expected"
fi
}
# ──────────────────────────────────────────────────────────────────────────────
# ─── DATE ARGUMENT ────────────────────────────────────────────────────────────
if [ "${1:-}" = "today" ]; then
TARGET_DATE=$(date +%Y-%m-%d)
elif [ -n "${1:-}" ]; then
TARGET_DATE="$1"
if ! [[ "$TARGET_DATE" =~ ^[0-9]{4}-[0-9]{2}-[0-9]{2}$ ]]; then
echo "ERROR: Invalid date format. Use YYYY-MM-DD, 'today', or no argument for yesterday."
exit 1
fi
else
TARGET_DATE=$(date -v-1d +%Y-%m-%d)
fi
log "========================================"
log "Screenpipe sync starting for: $TARGET_DATE"
log "========================================"
# ─── PREFLIGHT ────────────────────────────────────────────────────────────────
step "Preflight checks"
if [ ! -f "$DB_SRC" ]; then
log "ERROR: Source DB not found at $DB_SRC"; exit 1
fi
printf " %-20s %s (%s)\n" "Source DB:" "OK" "$(du -sh "$DB_SRC" | cut -f1)"
if [ ! -d "$NAS_MOUNT" ]; then
log "ERROR: NAS not mounted at $NAS_MOUNT"; exit 1
fi
printf " %-20s %s\n" "NAS mount:" "OK $NAS_MOUNT"
# Check if DB already synced for this date
DB_ALREADY_SYNCED=false
if [ -f "$NAS_DB" ]; then
EXISTING=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';" 2>/dev/null || echo "0")
if [ "$EXISTING" -gt "0" ]; then
log "Date $TARGET_DATE already has $EXISTING frames in archive — skipping DB sync"
DB_ALREADY_SYNCED=true
else
printf " %-20s %s (%s)\n" "Archive DB:" "exists" "$(du -sh "$NAS_DB" | cut -f1)"
fi
else
printf " %-20s %s\n" "Archive DB:" "will be created"
fi
# Source data dir for this date
DATA_SRC="$HOME/.screenpipe/data/data/$TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
DATA_SIZE=$(du -sh "$DATA_SRC" | cut -f1)
DATA_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
printf " %-20s %s (%s files, %s)\n" "Data dir:" "OK" "$DATA_FILES" "$DATA_SIZE"
else
printf " %-20s %s\n" "Data dir:" "not found — skipping file copy"
fi
# ─── DB SYNC ──────────────────────────────────────────────────────────────────
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── COUNT SOURCE ROWS ────────────────────────────────────────────────────
step "Counting source rows for $TARGET_DATE"
SRC_FRAMES=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
SRC_ELEMENTS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_UI=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
SRC_OCR=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_MEETINGS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start) = '$TARGET_DATE';")
printf " %-20s %s\n" "frames:" "$SRC_FRAMES"
printf " %-20s %s\n" "elements:" "$SRC_ELEMENTS"
printf " %-20s %s\n" "ui_events:" "$SRC_UI"
printf " %-20s %s\n" "ocr_text:" "$SRC_OCR"
printf " %-20s %s\n" "meetings:" "$SRC_MEETINGS"
if [ "$SRC_FRAMES" -eq "0" ]; then
log "No frames found for $TARGET_DATE — skipping DB sync"
DB_ALREADY_SYNCED=true
fi
fi
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── INIT TABLES ──────────────────────────────────────────────────────────
step "Initialising tables, indexes, FTS"
run_sqlite_heredoc "creating tables" "
ATTACH '$NAS_DB' AS nas;
CREATE TABLE IF NOT EXISTS nas.frames AS SELECT * FROM main.frames WHERE 0;
CREATE TABLE IF NOT EXISTS nas.elements AS SELECT * FROM main.elements WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ui_events AS SELECT * FROM main.ui_events WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ocr_text AS SELECT * FROM main.ocr_text WHERE 0;
CREATE TABLE IF NOT EXISTS nas.video_chunks AS SELECT * FROM main.video_chunks WHERE 0;
CREATE TABLE IF NOT EXISTS nas.meetings AS SELECT * FROM main.meetings WHERE 0;
DETACH nas;
"
run_sqlite_heredoc "creating indexes" "
ATTACH '$NAS_DB' AS nas;
CREATE INDEX IF NOT EXISTS nas.idx_frames_timestamp ON frames(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_frames_app_name ON frames(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_window_name ON frames(window_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_video_chunk_id ON frames(video_chunk_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_id ON elements(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_src_role ON elements(frame_id, source, role) WHERE text IS NOT NULL;
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_timestamp ON ui_events(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_app_name ON ui_events(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_frame_id ON ui_events(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_ocr_text_frame_id ON ocr_text(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_meetings_start ON meetings(meeting_start);
CREATE INDEX IF NOT EXISTS nas.idx_video_chunks_device ON video_chunks(device_name);
DETACH nas;
"
run_sqlite_heredoc "creating FTS tables" "
ATTACH '$NAS_DB' AS nas;
CREATE VIRTUAL TABLE IF NOT EXISTS nas.elements_fts USING fts5(
text, role, frame_id UNINDEXED,
content='elements', content_rowid='id', tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.frames_fts USING fts5(
full_text, app_name, window_name, browser_url, id UNINDEXED,
tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.ui_events_fts USING fts5(
text_content,
app_name,
window_title,
element_name,
content='ui_events',
content_rowid='id',
tokenize='unicode61'
);
DETACH nas;
"
# ─── SYNC DATA ────────────────────────────────────────────────────────────
step "Syncing data for $TARGET_DATE"
run_sqlite_heredoc "video_chunks" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.video_chunks
SELECT * FROM main.video_chunks
WHERE id IN (
SELECT DISTINCT video_chunk_id FROM main.frames
WHERE date(timestamp) = '$TARGET_DATE' AND video_chunk_id IS NOT NULL
);
DETACH nas;
"
run_sqlite_heredoc "frames ($SRC_FRAMES rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.frames
SELECT * FROM main.frames WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ocr_text ($SRC_OCR rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ocr_text
SELECT o.* FROM main.ocr_text o
JOIN main.frames f ON o.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ui_events ($SRC_UI rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ui_events
SELECT * FROM main.ui_events WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "elements ($SRC_ELEMENTS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.elements
SELECT e.* FROM main.elements e
JOIN main.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "meetings ($SRC_MEETINGS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.meetings
SELECT * FROM main.meetings WHERE date(meeting_start) = '$TARGET_DATE';
DETACH nas;
"
# ─── FTS UPDATE ───────────────────────────────────────────────────────────
step "Updating FTS indexes"
run_sqlite_heredoc "elements_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.elements_fts(rowid, text, role)
SELECT e.id, e.text, e.role
FROM nas.elements e
JOIN nas.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE'
AND e.text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "frames_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.frames_fts(rowid, full_text, app_name, window_name, browser_url, id)
SELECT id, full_text, app_name, window_name, browser_url, id
FROM nas.frames
WHERE date(timestamp) = '$TARGET_DATE'
AND full_text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "ui_events_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.ui_events_fts(rowid, text_content, app_name, window_title, element_name)
SELECT id, text_content, app_name, window_title, element_name
FROM nas.ui_events
WHERE date(timestamp) = '$TARGET_DATE'
AND text_content IS NOT NULL;
DETACH nas;
"
# ─── VERIFY DB ────────────────────────────────────────────────────────────
step "Verifying DB"
V_FRAMES=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
V_ELEMENTS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_UI=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
V_OCR=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_MEETINGS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start)= '$TARGET_DATE';")
check "frames" "$V_FRAMES" "$SRC_FRAMES"
check "elements" "$V_ELEMENTS" "$SRC_ELEMENTS"
check "ui_events" "$V_UI" "$SRC_UI"
check "ocr_text" "$V_OCR" "$SRC_OCR"
check "meetings" "$V_MEETINGS" "$SRC_MEETINGS"
fi
# ─── COPY DATA FOLDER ─────────────────────────────────────────────────────────
# Always runs regardless of DB sync status
step "Copying data folder for $TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
mkdir -p "$NAS_DATA/$TARGET_DATE"
RSYNC_START=$(date +%s)
printf " %-36s " "rsync $TARGET_DATE/ → NAS"
rsync -a --ignore-existing \
"$DATA_SRC/" \
"$NAS_DATA/$TARGET_DATE/" \
2>>"$LOG_FILE"
RSYNC_DUR=$(( $(date +%s) - RSYNC_START ))
COPIED_FILES=$(ls "$NAS_DATA/$TARGET_DATE" | wc -l | tr -d ' ')
SRC_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
COPIED_SIZE=$(du -sh "$NAS_DATA/$TARGET_DATE" | cut -f1)
if [ "$COPIED_FILES" -eq "$SRC_FILES" ]; then
printf "\r %-36s ✓ %dm%02ds (%s files, %s)\n" \
"rsync $TARGET_DATE/ → NAS" \
"$(( RSYNC_DUR / 60 ))" "$(( RSYNC_DUR % 60 ))" \
"$COPIED_FILES" "$COPIED_SIZE" | tee -a "$LOG_FILE"
else
printf "\r %-36s ✗ %s / %s files\n" \
"rsync $TARGET_DATE/ → NAS" "$COPIED_FILES" "$SRC_FILES" | tee -a "$LOG_FILE"
fi
else
printf " %-36s %s\n" "rsync $TARGET_DATE/ → NAS" "skipped (no source dir)"
fi
# ─── SUMMARY ──────────────────────────────────────────────────────────────────
TOTAL_ELAPSED=$(( $(date +%s) - SCRIPT_START ))
DB_SIZE=$(du -sh "$NAS_DB" | cut -f1)
echo ""
log "Archive DB size: $DB_SIZE"
log "Total time: $(( TOTAL_ELAPSED / 60 ))m$(( TOTAL_ELAPSED % 60 ))s"
log "Sync complete for $TARGET_DATE"
log "========================================"
#!/bin/bash
# screenpipe_sync.sh
# Syncs Screenpipe SQLite data to a NAS archive database (append-only, no deletions).
# Also copies the day's video/frame data folder to the NAS.
#
# Usage:
# ./screenpipe_sync.sh # syncs yesterday (default)
# ./screenpipe_sync.sh 2026-04-15 # syncs a specific date
# ./screenpipe_sync.sh today # syncs today so far
#
# Cron example (runs at 3am daily):
# 0 3 * * * /Users/lukas/.screenpipe/screenpipe_sync.sh >> /Users/lukas/.screenpipe/sync.log 2>&1
set -euo pipefail
# ─── CONFIG ───────────────────────────────────────────────────────────────────
DB_SRC="${SCREENPIPE_DB:-$HOME/.screenpipe/db.sqlite}"
NAS_MOUNT="${NAS_MOUNT:-/Volumes/Test/screenpipe}"
NAS_DB="$NAS_MOUNT/archive.db"
NAS_DATA="$NAS_MOUNT/data"
LOG_FILE="$HOME/.screenpipe/sync.log"
# ──────────────────────────────────────────────────────────────────────────────
# ─── HELPERS ──────────────────────────────────────────────────────────────────
SCRIPT_START=$(date +%s)
log() {
local msg="[$(date '+%Y-%m-%d %H:%M:%S')] $*"
echo "$msg" | tee -a "$LOG_FILE"
}
step() {
local now=$(date +%s)
local elapsed=$(( now - SCRIPT_START ))
local min=$(( elapsed / 60 ))
local sec=$(( elapsed % 60 ))
printf "\n[+%02dm%02ds] ▶ %s\n" "$min" "$sec" "$*" | tee -a "$LOG_FILE"
}
run_sqlite_heredoc() {
local label="$1"
local sql="$2"
local start=$(date +%s)
printf " %-36s " "$label"
sqlite3 "$DB_SRC" <<< "$sql" &
local pid=$!
local spin=[PASSWORD] '⠙' '⠹' '⠸' '⠼' '⠴' '⠦' '⠧' '⠇' '⠏')
local i=0
while kill -0 "$pid" 2>/dev/null; do
printf "\r %-36s %s " "$label" "${spin[$i]}"
i=$(( (i + 1) % 10 ))
sleep 0.2
done
wait "$pid"
local rc=$?
if [ $rc -ne 0 ]; then
printf "\r %-36s ✗ FAILED\n" "$label" | tee -a "$LOG_FILE"
exit $rc
fi
local dur=$(( $(date +%s) - start ))
printf "\r %-36s ✓ %dm%02ds\n" "$label" "$(( dur / 60 ))" "$(( dur % 60 ))" | tee -a "$LOG_FILE"
}
check() {
local label="$1" got="$2" expected="$3"
if [ "$got" -eq "$expected" ]; then
printf " %-20s %s / %s ✓\n" "$label:" "$got" "$expected"
else
printf " %-20s %s / %s ✗ MISMATCH\n" "$label:" "$got" "$expected"
fi
}
# ──────────────────────────────────────────────────────────────────────────────
# ─── DATE ARGUMENT ────────────────────────────────────────────────────────────
if [ "${1:-}" = "today" ]; then
TARGET_DATE=$(date +%Y-%m-%d)
elif [ -n "${1:-}" ]; then
TARGET_DATE="$1"
if ! [[ "$TARGET_DATE" =~ ^[0-9]{4}-[0-9]{2}-[0-9]{2}$ ]]; then
echo "ERROR: Invalid date format. Use YYYY-MM-DD, 'today', or no argument for yesterday."
exit 1
fi
else
TARGET_DATE=$(date -v-1d +%Y-%m-%d)
fi
log "========================================"
log "Screenpipe sync starting for: $TARGET_DATE"
log "========================================"
# ─── PREFLIGHT ────────────────────────────────────────────────────────────────
step "Preflight checks"
if [ ! -f "$DB_SRC" ]; then
log "ERROR: Source DB not found at $DB_SRC"; exit 1
fi
printf " %-20s %s (%s)\n" "Source DB:" "OK" "$(du -sh "$DB_SRC" | cut -f1)"
if [ ! -d "$NAS_MOUNT" ]; then
log "ERROR: NAS not mounted at $NAS_MOUNT"; exit 1
fi
printf " %-20s %s\n" "NAS mount:" "OK $NAS_MOUNT"
# Check if DB already synced for this date
DB_ALREADY_SYNCED=false
if [ -f "$NAS_DB" ]; then
EXISTING=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';" 2>/dev/null || echo "0")
if [ "$EXISTING" -gt "0" ]; then
log "Date $TARGET_DATE already has $EXISTING frames in archive — skipping DB sync"
DB_ALREADY_SYNCED=true
else
printf " %-20s %s (%s)\n" "Archive DB:" "exists" "$(du -sh "$NAS_DB" | cut -f1)"
fi
else
printf " %-20s %s\n" "Archive DB:" "will be created"
fi
# Source data dir for this date
DATA_SRC="$HOME/.screenpipe/data/data/$TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
DATA_SIZE=$(du -sh "$DATA_SRC" | cut -f1)
DATA_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
printf " %-20s %s (%s files, %s)\n" "Data dir:" "OK" "$DATA_FILES" "$DATA_SIZE"
else
printf " %-20s %s\n" "Data dir:" "not found — skipping file copy"
fi
# ─── DB SYNC ──────────────────────────────────────────────────────────────────
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── COUNT SOURCE ROWS ────────────────────────────────────────────────────
step "Counting source rows for $TARGET_DATE"
SRC_FRAMES=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
SRC_ELEMENTS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_UI=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
SRC_OCR=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_MEETINGS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start) = '$TARGET_DATE';")
printf " %-20s %s\n" "frames:" "$SRC_FRAMES"
printf " %-20s %s\n" "elements:" "$SRC_ELEMENTS"
printf " %-20s %s\n" "ui_events:" "$SRC_UI"
printf " %-20s %s\n" "ocr_text:" "$SRC_OCR"
printf " %-20s %s\n" "meetings:" "$SRC_MEETINGS"
if [ "$SRC_FRAMES" -eq "0" ]; then
log "No frames found for $TARGET_DATE — skipping DB sync"
DB_ALREADY_SYNCED=true
fi
fi
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── INIT TABLES ──────────────────────────────────────────────────────────
step "Initialising tables, indexes, FTS"
run_sqlite_heredoc "creating tables" "
ATTACH '$NAS_DB' AS nas;
CREATE TABLE IF NOT EXISTS nas.frames AS SELECT * FROM main.frames WHERE 0;
CREATE TABLE IF NOT EXISTS nas.elements AS SELECT * FROM main.elements WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ui_events AS SELECT * FROM main.ui_events WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ocr_text AS SELECT * FROM main.ocr_text WHERE 0;
CREATE TABLE IF NOT EXISTS nas.video_chunks AS SELECT * FROM main.video_chunks WHERE 0;
CREATE TABLE IF NOT EXISTS nas.meetings AS SELECT * FROM main.meetings WHERE 0;
DETACH nas;
"
run_sqlite_heredoc "creating indexes" "
ATTACH '$NAS_DB' AS nas;
CREATE INDEX IF NOT EXISTS nas.idx_frames_timestamp ON frames(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_frames_app_name ON frames(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_window_name ON frames(window_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_video_chunk_id ON frames(video_chunk_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_id ON elements(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_src_role ON elements(frame_id, source, role) WHERE text IS NOT NULL;
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_timestamp ON ui_events(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_app_name ON ui_events(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_frame_id ON ui_events(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_ocr_text_frame_id ON ocr_text(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_meetings_start ON meetings(meeting_start);
CREATE INDEX IF NOT EXISTS nas.idx_video_chunks_device ON video_chunks(device_name);
DETACH nas;
"
run_sqlite_heredoc "creating FTS tables" "
ATTACH '$NAS_DB' AS nas;
CREATE VIRTUAL TABLE IF NOT EXISTS nas.elements_fts USING fts5(
text, role, frame_id UNINDEXED,
content='elements', content_rowid='id', tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.frames_fts USING fts5(
full_text, app_name, window_name, browser_url, id UNINDEXED,
tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.ui_events_fts USING fts5(
text_content,
app_name,
window_title,
element_name,
content='ui_events',
content_rowid='id',
tokenize='unicode61'
);
DETACH nas;
"
# ─── SYNC DATA ────────────────────────────────────────────────────────────
step "Syncing data for $TARGET_DATE"
run_sqlite_heredoc "video_chunks" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.video_chunks
SELECT * FROM main.video_chunks
WHERE id IN (
SELECT DISTINCT video_chunk_id FROM main.frames
WHERE date(timestamp) = '$TARGET_DATE' AND video_chunk_id IS NOT NULL
);
DETACH nas;
"
run_sqlite_heredoc "frames ($SRC_FRAMES rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.frames
SELECT * FROM main.frames WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ocr_text ($SRC_OCR rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ocr_text
SELECT o.* FROM main.ocr_text o
JOIN main.frames f ON o.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ui_events ($SRC_UI rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ui_events
SELECT * FROM main.ui_events WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "elements ($SRC_ELEMENTS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.elements
SELECT e.* FROM main.elements e
JOIN main.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "meetings ($SRC_MEETINGS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.meetings
SELECT * FROM main.meetings WHERE date(meeting_start) = '$TARGET_DATE';
DETACH nas;
"
# ─── FTS UPDATE ───────────────────────────────────────────────────────────
step "Updating FTS indexes"
run_sqlite_heredoc "elements_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.elements_fts(rowid, text, role)
SELECT e.id, e.text, e.role
FROM nas.elements e
JOIN nas.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE'
AND e.text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "frames_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.frames_fts(rowid, full_text, app_name, window_name, browser_url, id)
SELECT id, full_text, app_name, window_name, browser_url, id
FROM nas.frames
WHERE date(timestamp) = '$TARGET_DATE'
AND full_text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "ui_events_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.ui_events_fts(rowid, text_content, app_name, window_title, element_name)
SELECT id, text_content, app_name, window_title, element_name
FROM nas.ui_events
WHERE date(timestamp) = '$TARGET_DATE'
AND text_content IS NOT NULL;
DETACH nas;
"
# ─── VERIFY DB ────────────────────────────────────────────────────────────
step "Verifying DB"
V_FRAMES=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
V_ELEMENTS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_UI=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
V_OCR=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_MEETINGS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start)= '$TARGET_DATE';")
check "frames" "$V_FRAMES" "$SRC_FRAMES"
check "elements" "$V_ELEMENTS" "$SRC_ELEMENTS"
check "ui_events" "$V_UI" "$SRC_UI"
check "ocr_text" "$V_OCR" "$SRC_OCR"
check "meetings" "$V_MEETINGS" "$SRC_MEETINGS"
fi
# ─── COPY DATA FOLDER ─────────────────────────────────────────────────────────
# Always runs regardless of DB sync status
step "Copying data folder for $TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
mkdir -p "$NAS_DATA/$TARGET_DATE"
RSYNC_START=$(date +%s)
printf " %-36s " "rsync $TARGET_DATE/ → NAS"
rsync -a --ignore-existing \
"$DATA_SRC/" \
"$NAS_DATA/$TARGET_DATE/" \
2>>"$LOG_FILE"
RSYNC_DUR=$(( $(date +%s) - RSYNC_START ))
COPIED_FILES=$(ls "$NAS_DATA/$TARGET_DATE" | wc -l | tr -d ' ')
SRC_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
COPIED_SIZE=$(du -sh "$NAS_DATA/$TARGET_DATE" | cut -f1)
if [ "$COPIED_FILES" -eq "$SRC_FILES" ]; then
printf "\r %-36s ✓ %dm%02ds (%s files, %s)\n" \
"rsync $TARGET_DATE/ → NAS" \
"$(( RSYNC_DUR / 60 ))" "$(( RSYNC_DUR % 60 ))" \
"$COPIED_FILES" "$COPIED_SIZE" | tee -a "$LOG_FILE"
else
printf "\r %-36s ✗ %s / %s files\n" \
"rsync $TARGET_DATE/ → NAS" "$COPIED_FILES" "$SRC_FILES" | tee -a "$LOG_FILE"
fi
else
printf " %-36s %s\n" "rsync $TARGET_DATE/ → NAS" "skipped (no source dir)"
fi
# ─── SUMMARY ──────────────────────────────────────────────────────────────────
TOTAL_ELAPSED=$(( $(date +%s) - SCRIPT_START ))
DB_SIZE=$(du -sh "$NAS_DB" | cut -f1)
echo ""
log "Archive DB size: $DB_SIZE"
log "Total time: $(( TOTAL_ELAPSED / 60 ))m$(( TOTAL_ELAPSED % 60 ))s"
log "Sync complete for $TARGET_DATE"
log "========================================"
lets add ability to jump…, Editor Group 2
Problems (⇧⌘M)
PROBLEMS
Output (⇧⌘U)
OUTPUT
Debug Console (⇧⌘Y)
DEBUG CONSOLE
Terminal (⌃`)
TERMINAL
Ports
PORTS
remote SSH: nas
SSH: nas
screenpipe (Git) - master*, Checkout Branch/Tag...
master*
screenpipe (Git) - Synchronize Changes
No Problems
0
0
No Ports Forwarded
0
Notifications
Signed out
Signed out
Shell Script
Editor Language Status: $(copilot) No inline suggestion available, Inline suggestions
LF
UTF-8
Spaces: 2
Ln 320, Col 36
Screen Reader Optimized
git-commit Lukas Kovalik (23 hours ago)
Lukas Kovalik (23 hours ago)
Info: Downloading VS Code Server...
lets add ability to jump fro,m search to timeline so we can see the exact screen (also based on monitor and time
Session history
New session
Message actions
This session is being continued from a previous conversation that ran out of context. The summary below covers the earlier portion of the conversation.
Summary:
1. Primary Request and Intent:
The user has been developing a Screenpipe dashboard on a Synology NAS. Requests in chronological order:
- **Q1**: Are db.sqlite-shm and db.sqlite-wal used? → Explained WAL mode; cleaned up files.
- **Q2**: AND combination search (term1 + term2 must both appear) → Added "AND also" input field.
- **Q3**: Explain UTC time issue, show date in results, make Only/Skip apps use multiselect → Implemented all three.
- **Q4**: Point sqlite-web at archive.db, add second sqlite-web for db.sqlite → Updated docker-compose.yml, added two services on ports 8767/8768.
- **Q5**: "still the same" (container not picking up config) → Explained `up -d` vs `restart`.
- **Q6**: Add git, push to gitea.com/lakylak/screenpipe.git → Initialized repo, committed, pushed.
- **Q7**: Add Timetable tab with video playback and timeline → Full implementation with click-to-seek, auto-advancing chunks, legend.
- **Q8**: Add scroll/zoom on timeline, jump to specific time, better awareness → Implemented zoom (1×–32×), scrollable timeline, overview minimap, viewport indicator, Jump-to-time input, screenshot thumbnail.
- **Q9 (most recent)**: Make playing speed more than 8×, fix speed not working, show position on timeline → CSS fix done (wider playhead + time label style), remainder in progress.
2. Key Technical Concepts:
- **SQLite WAL mode**: db.sqlite-shm and db.sqlite-wal are WAL companion files; cleaned with `PRAGMA wal_checkpoint(TRUNCATE)`.
- **SQLite FTS5**: Space-separated terms = AND; `MATCH 'term1 term2'` requires both; OR requires explicit `OR` keyword.
- **FastAPI**: Python backend with route handlers in main.py, consumers pattern for DB queries.
- **Docker Compose**: App image is built (code baked in), not volume-mounted → requires `build` + `up -d` for code changes. `restart` doesn't re-read compose file — need `up -d`.
- **sqlite-web**: Read-only flag `-r`, pre-built image, no rebuild needed for config changes.
- **Video playback**: screenpipe records at 0.1–0.3 fps (timelapse). `seek = offset_index / fps` gives seconds into video chunk. ~245 chunks/day across 2 monitors. Videos may be encoded at standard fps making them very short at native speed.
- **Timeline rendering**: App-colored segments built by merging consecutive frames with same `app_name`. Percentage positions: `((ts - dayStart) / span * 100)%`.
- **Zoom/scroll**: Inner div width = `zoom × 100%`, outer wrapper has `overflow-x:auto`. Playhead position stays as % of inner div, which is correct at all zoom levels.
- **Timer-based speed**: For speed > 2×, browser `playbackRate` is unreliable on low-FPS video; correct approach is `setInterval` advancing `video.currentTime` by `speed * interval_ms / 1000` seconds per tick.
- **UTC vs local time**: Backend treats `time_from`/`time_to` as UTC. Frontend converts with `localTimeToUtc()` before sending.
- **Git**: `#recycle/` in .gitignore treated as comment — must escape as `\#recycle/`.
3. Files and Code Sections:
- **`/volume1/screenpipe/app/main.py`**
- Added `/api/apps` endpoint (all unique app names from frames table)
- Added `/api/timetable` endpoint (frames + video URLs + screenshot URLs per device per date)
- `/api/timetable` response structure: `{frames: [{ts, app, win, video_url, seek, screenshot_url}], devices: [...], device: "..."}`
- Key snippet:
```python
@app.get("/api/timetable"
)
async def timetable_data(d: str | None = Query(None), device: str | None = Query(None)):
...
rows = conn.execute(f"""
SELECT f.timestamp AS ts, f.app_name AS app, f.window_name AS win,
f.offset_index, f.snapshot_path,
vc.file_path AS video_path, vc.fps AS video_fps
FROM frames f
JOIN video_chunks vc ON f.video_chunk_id = vc.id
WHERE f.timestamp >= ? AND f.timestamp < ?
AND f.video_chunk_id IS NOT NULL {dev_clause}
ORDER BY f.timestamp
""", params).fetchall()
# seek = offset_index / fps, screenshot_url from _snapshot_to_url(snapshot_path)
```
- **`/volume1/screenpipe/app/docker-compose.yml`**
- Changed sqlite-web service name to `sqlite-web-archive` → port 8767 → `/data/archive.db`
- Added `sqlite-web-live` → port 8768 → `/data/db.sqlite`
- screenpipe-app: port 8766, `DB_PATH: /data/db.sqlite`, `TZ: Europe/Sofia`
- **`/volume1/screenpipe/app/static/index.html`** (single large file with all CSS/HTML/JS)
- **Search enhancements**:
- Added `id="sq2"` "AND also" input between search-row1 and search-row2
- Replaced `sOnlyApps`/`sSkipApps` text inputs with custom multiselect components (`id="msOnlyApps"`, `id="msSkipApps"`)
- Updated `buildUrl()` to combine `q + " " + q2` and convert times with `localTimeToUtc()`
- Added `fmtDateTime(ts)` showing "Apr 20 14:08" format
- Added `localTimeToUtc(timeStr)` converting local HH:MM to UTC
- Multiselect state: `const _msState = { msOnlyApps: new Set(), msSkipApps: new Set() }`
- **Timetable tab** (new page):
- Tab button added between Work Report and AI Summary
- `showPage` array updated: `['activity','search','audio','report','timetable','summary']`
- `refresh()` calls `loadTimetable()` when on timetable page
- **Timetable CSS** (key styles):
```css
.tt-scroll-wrap { overflow-x:auto; ... scrollbar-width:thin; }
.tt-inner { position:relative; height:52px; } /* width set dynamically */
.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
.tt-overview-wrap { position:relative; height:14px; ... cursor:pointer; }
.tt-overview-vp { position:absolute; background:rgba(255,255,255,0.12); border:1px solid rgba(255,255,255,0.4); }
```
- **Timetable JS state variables**:
```javascript
let _ttFrames = [], _ttChunks = [], _ttCurChunk = -1;
let _ttDayStart = 0, _ttDayEnd = 0, _ttZoom = 1;
const _ttAppColors = {};
const TT_ZOOMS = [1, 2, 4, 8, 16, 32];
const TT_TICK_MIN = {1:30, 2:15, 4:10, 8:5, 16:2, 32:1};
```
- **Key JS functions**:
- `loadTimetable()` — fetches `/api/timetable`, groups chunks, renders
- `ttZoom(dir)` — changes zoom level, updates inner width, re-renders ticks, scrolls to playhead
- `ttTimelineClick(e)` — uses `inner.getBoundingClientRect()` (works correctly when scrolled)
- `ttOverviewClick(e)` — scrolls timeline + seeks video
- `ttUpdateOverviewVp()` — updates white viewport rect using `wrap.scrollLeft / inner.scrollWidth`
- `ttJumpToTime()` — parses time input, converts to ms, calls `ttSeekToMs`
- `ttSeekToMs(tsMs)` — binary-searches nearest frame, loads chunk, seeks video
- `_ttScrollToPlayhead(force)` — scrolls wrap so playhead stays at ~35% from left when Follow is checked
- `_ttMovePlayhead(tsMs)` — sets `ph.style.left` as percentage, calls `_ttScrollToPlayhead`
- `_ttShowNow(f)` — updates screenshot thumbnail + now-text div
- `ttSyncBtn()` — syncs play/pause button text
- Keyboard: Space=play/pause, ←/→=±10s, Ctrl+←/→=±30s, +/-=zoom
- **`/volume1/screenpipe/.gitignore`**
```
*.db, *.sqlite, *.sqlite-shm, *.sqlite-wal
data/, \#recycle/,
@eaDir/
app_settings.json, app/.env
__pycache__/, *.pyc, .DS_Store
```
4. Errors and Fixes:
- **`#recycle/` in .gitignore treated as comment**: `#` starts a comment in .gitignore. Fixed by escaping: `\#recycle/`. Also had to `git rm --cached "#recycle/desktop.ini"` to unstage it.
- **`sudo docker compose restart sqlite-web` didn't apply config change**: `restart` reuses existing container config, doesn't re-read compose file. Fix: `sudo docker compose up -d` recreates containers with new config.
- **Orphan container holding port 8767**: Old `screenpipe-sqlweb` container still running after rename. Fix: `sudo docker compose up -d --remove-orphans`.
- **`sudo docker` not accessible from Claude Code**: Interactive sudo required. User runs commands in NAS terminal.
- **`git config --global --add safe.directory /volume1/screenpipe`**: Required due to dubious ownership of NAS volume directory.
5. Problem Solving:
- **Time zone confusion in search**: Backend uses UTC for time_from/time_to; frontend displays local time. Solved by `localTimeToUtc()` converting before API call, and showing date+time in results.
- **sqlite-web not showing April 17–21 data**: Container was pointing at db.sqlite (old config), not archive.db. Fixed by updating docker-compose.yml + `up -d`.
- **Video playback speed**: Browser `video.playbackRate` is capped (~16× max in Chrome) and performs poorly for very low FPS (0.1–0.3 fps) screenpipe videos. Each chunk may only be 1–2 seconds of wall-clock encoded video, playing through instantly. Solution: implement timer-based frame-jumping for speed > 2× (in progress).
- **Playhead visibility**: White 2px line was hard to see. Fixed CSS to 3px width + stronger box-shadow + added `.tt-playhead-time` overlay label style.
6. All User Messages:
- "do we use db.sqlite-shm and db.sqlite-wal for anything right now? [image of search UI] Lets add ability to search with combination... Explain how does the search work time seems to work in utc maybe display date next to the time in results. Make only apps and skip apps multiselect with all available apps."
- "doi I need to restart?"
- "there is sudo docker compose"
- "no still the same"
- "add git. I have created gitea repo https://gitea.com/lakylak/screenpipe.git"
- "just do it" (re: git push)
- "how to make sure data in :8767 are accurate and point to archive.db"
- "I can't see older than 16th there should 21st of april. In http://[IP_ADDRESS]:8766/ I can see all but not on sqlite viewer http://[IP_ADDRESS]:8767"
- "ok lets add additional port to read db.sqlite, I would like to see the structure of official db that reads db.sqlite db and one for archive"
- [docker compose output showing orphan container error]
- "looks great. I want to have ability to scroll left and right and also to jump to specific time. I want to be more aware to what I see. Still it should be fast"
- "lets make the playing speed more than 8x"
- "still playing speed is not showing properly still just in one second. also I want to see in timeline where I am"
7. Pending Tasks:
- **Playback speed fix** (in progress, partially done):
- Add `<div class="tt-playhead-time" id="ttPlayheadTime"></div>` inside `ttInner` HTML
- Implement `_ttStartSpeedTimer(speed)` / `_ttStopSpeedTimer()` / `ttSetSpeed(speed)` functions
- Update speed select `onchange` to call `ttSetSpeed(+this.value)`
- Add 64× option to speed select
- Update `_ttMovePlayhead` to also update `ttPlayheadTime` label
- Update `ttSyncBtn` to start/stop timer based on play state + speed
- Update `ttNextChunk` and `ttSeekToMs` to stop timer (restarted by onplay → ttSyncBtn)
- Rebuild Docker image after changes
8. Current Work:
Working on fixing two issues with the Timetable tab:
1. **Playback speed**: `video.playbackRate` doesn't work well for low-FPS screenpipe videos. The plan is to implement a `setInterval`-based frame-jumper for speed > 2× that advances `video.currentTime` by `speed * intervalMs / 1000` every 100–200ms, bypassing browser playbackRate limits.
2. **Timeline position indicator**: Playhead not prominent enough.
**CSS already changed** in `/volume1/screenpipe/app/static/index.html`:
```css
.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
```
Session was interrupted by summary request after this CSS edit, mid-implementation.
9. Optional Next Step:
Continuing the fix for: "still playing speed is not showing properly still just in one second. also I want to see in timeline where I am"
The next steps in order are:
1. Add `<div class="tt-playhead-time" id="ttPlayheadTime"></div>` to the HTML inside `#ttInner` (after the existing `.tt-playhead` div)
2. Update `_ttMovePlayhead` to set `ttPlayheadTime` position and text content with `fmtTime()`
3. Add timer-based speed system to JS (new functions `_ttStartSpeedTimer`, `_ttStopSpeedTimer`, `ttSetSpeed`)
4. Update speed select's `onchange` to `ttSetSpeed(+this.value)` and add 64× option
5. Update `ttSyncBtn`, `ttNextChunk`, `ttSeekToMs` to integrate with timer
6. Rebuild: `sudo docker compose build screenpipe-app && sudo docker compose up -d screenpipe-app`
If you need specific details from before compaction (like exact code snippets, error messages, or content you generated), read the full transcript at: /home/Adm1n/.claude/projects/-volume1-screenpipe/7777f46e-1e40-4051-b4bb-f43aedd59556.jsonl
Continue the conversation from where it left off without asking the user any further questions. Resume directly — do not acknowledge the summary, do not recap what was happening, do not preface with "I'll continue" or similar. Pick up the last task as if the break never happened.
Thinking
Thinking
Bash
Find timetable speed and playhead related code
IN
grep -n "tt-playhead\|ttPlayhead\|ttInner\|_ttMovePlayhead\|ttSetSpeed\|_ttStartSpeed\|_ttStopSpeed\|ttSyncBtn\|playbackRate\|speed.*select\|spdSel\|ttSpeed" /volume1/screenpipe/app/static/index.html | head -80
Copy code to clipboard
OUT
207:.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
208:.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
580: <div class="tt-inner" id="ttInner" onclick="ttTimelineClick(event)">
583: <div class="tt-playhead" id="ttPlayhead"></div>
598: onpause="ttSyncBtn()"
599: onplay="ttSyncBtn()">
607: <select id="ttSpeed" onchange="document.getElementById('ttVideo').playbackRate=+this.value">
1624: document.getElementById('ttInner').style.width = '100%';
1632: document.getElementById('ttPlayhead').style.display = 'none';
1694: document.getElementById('ttInner').style.width = (_ttZoom * 100) + '%';
1700: const inner = document.getElementById('ttInner');
1710: const inner = document.getElementById('ttInner');
1718: const inner = document.getElementById('ttInner');
1750: _ttMovePlayhead(new Date(best.ts).getTime());
1772: _ttMovePlayhead(new Date(best.ts).getTime());
1791:function _ttMovePlayhead(tsMs) {
1792: const ph = document.getElementById('ttPlayhead');
1802: const ph = document.getElementById('ttPlayhead');
1805: const inner = document.getElementById('ttInner');
1819:function ttSyncBtn() {
Read
index.html
index.html
(lines 576-625)
Read
index.html
index.html
(lines 1786-1865)
Thinking
Thinking
Bash
Find chunk loading and playhead update functions
IN
grep -n "ttUpdatePlayhead\|ttNextChunk\|ttLoadChunk\|_ttCurChunk\|_ttChunks\|_ttFrames" /volume1/screenpipe/app/static/index.html | head -50
Copy code to clipboard
OUT
596: onended="ttNextChunk()"
597: ontimeupdate="ttUpdatePlayhead()"
1569:let _ttFrames = [];
1570:let _ttChunks = [];
1571:let _ttCurChunk = -1;
1607: _ttFrames = data.frames;
1609: for (const f of _ttFrames) {
1613: _ttChunks = [...chunkMap.values()];
1614: _ttCurChunk = -1;
1616: _ttDayStart = new Date(_ttFrames[0].ts).getTime();
1617: _ttDayEnd = new Date(_ttFrames[_ttFrames.length - 1].ts).getTime();
1620: for (const f of _ttFrames) _ttColor(f.app);
1640: for (const f of _ttFrames) {
1736: if (!_ttFrames.length) return;
1737: let best = _ttFrames[0], minDiff = Infinity;
1738: for (const f of _ttFrames) {
1742: const ci = _ttChunks.findIndex(ch => ch[0].video_url === best.video_url);
1744: _ttCurChunk = ci;
1753:function ttNextChunk() {
1754: if (_ttCurChunk + 1 >= _ttChunks.length) return;
1755: _ttCurChunk++;
1757: video.src = _ttChunks[_ttCurChunk][0].video_url;
1762:function ttUpdatePlayhead() {
1764: if (_ttCurChunk < 0 || _ttCurChunk >= _ttChunks.length) return;
1765: const chunk = _ttChunks[_ttCurChunk];
Read
index.html
index.html
(lines 1754-1793)
Thinking
Thinking
Read
index.html
index.html
(lines 1701-1760)
Thinking
Thinking
Read
index.html
index.html
(lines 1561-1640)
Thinking
Thinking
Bash
Find time formatting functions
IN
grep -n "function fmtTime\|function fmtDate\|function _fmt\|localTimeToUtc\|fmtDateTime" /volume1/screenpipe/app/static/index.html | head -20
Copy code to clipboard
OUT
1088: if (tf) url += `&time_from=${localTimeToUtc(tf)}`;
1089: if (tt) url += `&time_to=${localTimeToUtc(tt)}`;
1153: let meta = `<span style="color:var(--muted)">${fmtDateTime(r.timestamp)}</span>${srcBadge}`;
1166: meta = `<span style="color:var(--muted)">${fmtDateTime(r.timestamp)}</span>${srcBadge} <span class="badge ${r.is_input_device?'mic':'sys'}">${r.is_input_device?'MIC':'SYS'}</span>`;
1307:function fmtTime(ts) {
1312:function fmtDateTime(ts) {
1320:function localTimeToUtc(timeStr) {
1778:function _fmtSec(s) {
1788: `${fmtDateTime(f.ts)} · ${f.app}${f.win ? ' / ' + f.win : ''}`;
Read
index.html
index.html
(lines 1308-1327)
Thinking
Thinking
Now I have everything I need. I'll implement the frame-by-frame timer system that replaces
video.playbackRate
with wall-clock accurate playback, plus add the playhead time label. Let me make the changes:
Edit
index.html
index.html
Added 1 line...
|
Code
|
screenpipe_sync.sh — screenpipe [SSH: nas]
|
NULL
|
|
Explorer (⇧⌘E)
Search (⇧⌘F)
Source Control (⌃⇧ Explorer (⇧⌘E)
Search (⇧⌘F)
Source Control (⌃⇧G) - 2 pending changes
2
Run and Debug (⇧⌘D)
Remote Explorer
Extensions (⇧⌘X) - 2 require update, 1 requires restart
3
Claude Code
Containers
EXPLORER
EXPLORER
Explorer Section: screenpipe [SSH: nas]
Explorer Section: screenpipe [SSH: nas]
SCREENPIPE [SSH: NAS]
#recycle
app
data
pipes
.gitignore
app_settings.json
archive.db
db.sqlite
screenpipe_sync.sh
screenpipe.db
Outline Section
OUTLINE
OUTLINE
Timeline Section
TIMELINE
TIMELINE
app_settings.json, Editor Group 1
screenpipe_sync.sh, Editor Group 1
db.sqlite-shm, preview, Editor Group 1
#!/bin/bash
# screenpipe_sync.sh
# Syncs Screenpipe SQLite data to a NAS archive database (append-only, no deletions).
# Also copies the day's video/frame data folder to the NAS.
#
# Usage:
# ./screenpipe_sync.sh # syncs yesterday (default)
# ./screenpipe_sync.sh 2026-04-15 # syncs a specific date
# ./screenpipe_sync.sh today # syncs today so far
#
# Cron example (runs at 3am daily):
# 0 3 * * * /Users/lukas/.screenpipe/screenpipe_sync.sh >> /Users/lukas/.screenpipe/sync.log 2>&1
set -euo pipefail
# ─── CONFIG ───────────────────────────────────────────────────────────────────
DB_SRC="${SCREENPIPE_DB:-$HOME/.screenpipe/db.sqlite}"
NAS_MOUNT="${NAS_MOUNT:-/Volumes/Test/screenpipe}"
NAS_DB="$NAS_MOUNT/archive.db"
NAS_DATA="$NAS_MOUNT/data"
LOG_FILE="$HOME/.screenpipe/sync.log"
# ──────────────────────────────────────────────────────────────────────────────
# ─── HELPERS ──────────────────────────────────────────────────────────────────
SCRIPT_START=$(date +%s)
log() {
local msg="[$(date '+%Y-%m-%d %H:%M:%S')] $*"
echo "$msg" | tee -a "$LOG_FILE"
}
step() {
local now=$(date +%s)
local elapsed=$(( now - SCRIPT_START ))
local min=$(( elapsed / 60 ))
local sec=$(( elapsed % 60 ))
printf "\n[+%02dm%02ds] ▶ %s\n" "$min" "$sec" "$*" | tee -a "$LOG_FILE"
}
run_sqlite_heredoc() {
local label="$1"
local sql="$2"
local start=$(date +%s)
printf " %-36s " "$label"
sqlite3 "$DB_SRC" <<< "$sql" &
local pid=$!
local spin=[PASSWORD] '⠙' '⠹' '⠸' '⠼' '⠴' '⠦' '⠧' '⠇' '⠏')
local i=0
while kill -0 "$pid" 2>/dev/null; do
printf "\r %-36s %s " "$label" "${spin[$i]}"
i=$(( (i + 1) % 10 ))
sleep 0.2
done
wait "$pid"
local rc=$?
if [ $rc -ne 0 ]; then
printf "\r %-36s ✗ FAILED\n" "$label" | tee -a "$LOG_FILE"
exit $rc
fi
local dur=$(( $(date +%s) - start ))
printf "\r %-36s ✓ %dm%02ds\n" "$label" "$(( dur / 60 ))" "$(( dur % 60 ))" | tee -a "$LOG_FILE"
}
check() {
local label="$1" got="$2" expected="$3"
if [ "$got" -eq "$expected" ]; then
printf " %-20s %s / %s ✓\n" "$label:" "$got" "$expected"
else
printf " %-20s %s / %s ✗ MISMATCH\n" "$label:" "$got" "$expected"
fi
}
# ──────────────────────────────────────────────────────────────────────────────
# ─── DATE ARGUMENT ────────────────────────────────────────────────────────────
if [ "${1:-}" = "today" ]; then
TARGET_DATE=$(date +%Y-%m-%d)
elif [ -n "${1:-}" ]; then
TARGET_DATE="$1"
if ! [[ "$TARGET_DATE" =~ ^[0-9]{4}-[0-9]{2}-[0-9]{2}$ ]]; then
echo "ERROR: Invalid date format. Use YYYY-MM-DD, 'today', or no argument for yesterday."
exit 1
fi
else
TARGET_DATE=$(date -v-1d +%Y-%m-%d)
fi
log "========================================"
log "Screenpipe sync starting for: $TARGET_DATE"
log "========================================"
# ─── PREFLIGHT ────────────────────────────────────────────────────────────────
step "Preflight checks"
if [ ! -f "$DB_SRC" ]; then
log "ERROR: Source DB not found at $DB_SRC"; exit 1
fi
printf " %-20s %s (%s)\n" "Source DB:" "OK" "$(du -sh "$DB_SRC" | cut -f1)"
if [ ! -d "$NAS_MOUNT" ]; then
log "ERROR: NAS not mounted at $NAS_MOUNT"; exit 1
fi
printf " %-20s %s\n" "NAS mount:" "OK $NAS_MOUNT"
# Check if DB already synced for this date
DB_ALREADY_SYNCED=false
if [ -f "$NAS_DB" ]; then
EXISTING=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';" 2>/dev/null || echo "0")
if [ "$EXISTING" -gt "0" ]; then
log "Date $TARGET_DATE already has $EXISTING frames in archive — skipping DB sync"
DB_ALREADY_SYNCED=true
else
printf " %-20s %s (%s)\n" "Archive DB:" "exists" "$(du -sh "$NAS_DB" | cut -f1)"
fi
else
printf " %-20s %s\n" "Archive DB:" "will be created"
fi
# Source data dir for this date
DATA_SRC="$HOME/.screenpipe/data/data/$TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
DATA_SIZE=$(du -sh "$DATA_SRC" | cut -f1)
DATA_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
printf " %-20s %s (%s files, %s)\n" "Data dir:" "OK" "$DATA_FILES" "$DATA_SIZE"
else
printf " %-20s %s\n" "Data dir:" "not found — skipping file copy"
fi
# ─── DB SYNC ──────────────────────────────────────────────────────────────────
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── COUNT SOURCE ROWS ────────────────────────────────────────────────────
step "Counting source rows for $TARGET_DATE"
SRC_FRAMES=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
SRC_ELEMENTS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_UI=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
SRC_OCR=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_MEETINGS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start) = '$TARGET_DATE';")
printf " %-20s %s\n" "frames:" "$SRC_FRAMES"
printf " %-20s %s\n" "elements:" "$SRC_ELEMENTS"
printf " %-20s %s\n" "ui_events:" "$SRC_UI"
printf " %-20s %s\n" "ocr_text:" "$SRC_OCR"
printf " %-20s %s\n" "meetings:" "$SRC_MEETINGS"
if [ "$SRC_FRAMES" -eq "0" ]; then
log "No frames found for $TARGET_DATE — skipping DB sync"
DB_ALREADY_SYNCED=true
fi
fi
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── INIT TABLES ──────────────────────────────────────────────────────────
step "Initialising tables, indexes, FTS"
run_sqlite_heredoc "creating tables" "
ATTACH '$NAS_DB' AS nas;
CREATE TABLE IF NOT EXISTS nas.frames AS SELECT * FROM main.frames WHERE 0;
CREATE TABLE IF NOT EXISTS nas.elements AS SELECT * FROM main.elements WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ui_events AS SELECT * FROM main.ui_events WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ocr_text AS SELECT * FROM main.ocr_text WHERE 0;
CREATE TABLE IF NOT EXISTS nas.video_chunks AS SELECT * FROM main.video_chunks WHERE 0;
CREATE TABLE IF NOT EXISTS nas.meetings AS SELECT * FROM main.meetings WHERE 0;
DETACH nas;
"
run_sqlite_heredoc "creating indexes" "
ATTACH '$NAS_DB' AS nas;
CREATE INDEX IF NOT EXISTS nas.idx_frames_timestamp ON frames(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_frames_app_name ON frames(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_window_name ON frames(window_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_video_chunk_id ON frames(video_chunk_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_id ON elements(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_src_role ON elements(frame_id, source, role) WHERE text IS NOT NULL;
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_timestamp ON ui_events(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_app_name ON ui_events(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_frame_id ON ui_events(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_ocr_text_frame_id ON ocr_text(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_meetings_start ON meetings(meeting_start);
CREATE INDEX IF NOT EXISTS nas.idx_video_chunks_device ON video_chunks(device_name);
DETACH nas;
"
run_sqlite_heredoc "creating FTS tables" "
ATTACH '$NAS_DB' AS nas;
CREATE VIRTUAL TABLE IF NOT EXISTS nas.elements_fts USING fts5(
text, role, frame_id UNINDEXED,
content='elements', content_rowid='id', tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.frames_fts USING fts5(
full_text, app_name, window_name, browser_url, id UNINDEXED,
tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.ui_events_fts USING fts5(
text_content,
app_name,
window_title,
element_name,
content='ui_events',
content_rowid='id',
tokenize='unicode61'
);
DETACH nas;
"
# ─── SYNC DATA ────────────────────────────────────────────────────────────
step "Syncing data for $TARGET_DATE"
run_sqlite_heredoc "video_chunks" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.video_chunks
SELECT * FROM main.video_chunks
WHERE id IN (
SELECT DISTINCT video_chunk_id FROM main.frames
WHERE date(timestamp) = '$TARGET_DATE' AND video_chunk_id IS NOT NULL
);
DETACH nas;
"
run_sqlite_heredoc "frames ($SRC_FRAMES rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.frames
SELECT * FROM main.frames WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ocr_text ($SRC_OCR rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ocr_text
SELECT o.* FROM main.ocr_text o
JOIN main.frames f ON o.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ui_events ($SRC_UI rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ui_events
SELECT * FROM main.ui_events WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "elements ($SRC_ELEMENTS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.elements
SELECT e.* FROM main.elements e
JOIN main.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "meetings ($SRC_MEETINGS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.meetings
SELECT * FROM main.meetings WHERE date(meeting_start) = '$TARGET_DATE';
DETACH nas;
"
# ─── FTS UPDATE ───────────────────────────────────────────────────────────
step "Updating FTS indexes"
run_sqlite_heredoc "elements_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.elements_fts(rowid, text, role)
SELECT e.id, e.text, e.role
FROM nas.elements e
JOIN nas.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE'
AND e.text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "frames_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.frames_fts(rowid, full_text, app_name, window_name, browser_url, id)
SELECT id, full_text, app_name, window_name, browser_url, id
FROM nas.frames
WHERE date(timestamp) = '$TARGET_DATE'
AND full_text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "ui_events_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.ui_events_fts(rowid, text_content, app_name, window_title, element_name)
SELECT id, text_content, app_name, window_title, element_name
FROM nas.ui_events
WHERE date(timestamp) = '$TARGET_DATE'
AND text_content IS NOT NULL;
DETACH nas;
"
# ─── VERIFY DB ────────────────────────────────────────────────────────────
step "Verifying DB"
V_FRAMES=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
V_ELEMENTS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_UI=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
V_OCR=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_MEETINGS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start)= '$TARGET_DATE';")
check "frames" "$V_FRAMES" "$SRC_FRAMES"
check "elements" "$V_ELEMENTS" "$SRC_ELEMENTS"
check "ui_events" "$V_UI" "$SRC_UI"
check "ocr_text" "$V_OCR" "$SRC_OCR"
check "meetings" "$V_MEETINGS" "$SRC_MEETINGS"
fi
# ─── COPY DATA FOLDER ─────────────────────────────────────────────────────────
# Always runs regardless of DB sync status
step "Copying data folder for $TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
mkdir -p "$NAS_DATA/$TARGET_DATE"
RSYNC_START=$(date +%s)
printf " %-36s " "rsync $TARGET_DATE/ → NAS"
rsync -a --ignore-existing \
"$DATA_SRC/" \
"$NAS_DATA/$TARGET_DATE/" \
2>>"$LOG_FILE"
RSYNC_DUR=$(( $(date +%s) - RSYNC_START ))
COPIED_FILES=$(ls "$NAS_DATA/$TARGET_DATE" | wc -l | tr -d ' ')
SRC_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
COPIED_SIZE=$(du -sh "$NAS_DATA/$TARGET_DATE" | cut -f1)
if [ "$COPIED_FILES" -eq "$SRC_FILES" ]; then
printf "\r %-36s ✓ %dm%02ds (%s files, %s)\n" \
"rsync $TARGET_DATE/ → NAS" \
"$(( RSYNC_DUR / 60 ))" "$(( RSYNC_DUR % 60 ))" \
"$COPIED_FILES" "$COPIED_SIZE" | tee -a "$LOG_FILE"
else
printf "\r %-36s ✗ %s / %s files\n" \
"rsync $TARGET_DATE/ → NAS" "$COPIED_FILES" "$SRC_FILES" | tee -a "$LOG_FILE"
fi
else
printf " %-36s %s\n" "rsync $TARGET_DATE/ → NAS" "skipped (no source dir)"
fi
# ─── SUMMARY ──────────────────────────────────────────────────────────────────
TOTAL_ELAPSED=$(( $(date +%s) - SCRIPT_START ))
DB_SIZE=$(du -sh "$NAS_DB" | cut -f1)
echo ""
log "Archive DB size: $DB_SIZE"
log "Total time: $(( TOTAL_ELAPSED / 60 ))m$(( TOTAL_ELAPSED % 60 ))s"
log "Sync complete for $TARGET_DATE"
log "========================================"
#!/bin/bash
# screenpipe_sync.sh
# Syncs Screenpipe SQLite data to a NAS archive database (append-only, no deletions).
# Also copies the day's video/frame data folder to the NAS.
#
# Usage:
# ./screenpipe_sync.sh # syncs yesterday (default)
# ./screenpipe_sync.sh 2026-04-15 # syncs a specific date
# ./screenpipe_sync.sh today # syncs today so far
#
# Cron example (runs at 3am daily):
# 0 3 * * * /Users/lukas/.screenpipe/screenpipe_sync.sh >> /Users/lukas/.screenpipe/sync.log 2>&1
set -euo pipefail
# ─── CONFIG ───────────────────────────────────────────────────────────────────
DB_SRC="${SCREENPIPE_DB:-$HOME/.screenpipe/db.sqlite}"
NAS_MOUNT="${NAS_MOUNT:-/Volumes/Test/screenpipe}"
NAS_DB="$NAS_MOUNT/archive.db"
NAS_DATA="$NAS_MOUNT/data"
LOG_FILE="$HOME/.screenpipe/sync.log"
# ──────────────────────────────────────────────────────────────────────────────
# ─── HELPERS ──────────────────────────────────────────────────────────────────
SCRIPT_START=$(date +%s)
log() {
local msg="[$(date '+%Y-%m-%d %H:%M:%S')] $*"
echo "$msg" | tee -a "$LOG_FILE"
}
step() {
local now=$(date +%s)
local elapsed=$(( now - SCRIPT_START ))
local min=$(( elapsed / 60 ))
local sec=$(( elapsed % 60 ))
printf "\n[+%02dm%02ds] ▶ %s\n" "$min" "$sec" "$*" | tee -a "$LOG_FILE"
}
run_sqlite_heredoc() {
local label="$1"
local sql="$2"
local start=$(date +%s)
printf " %-36s " "$label"
sqlite3 "$DB_SRC" <<< "$sql" &
local pid=$!
local spin=[PASSWORD] '⠙' '⠹' '⠸' '⠼' '⠴' '⠦' '⠧' '⠇' '⠏')
local i=0
while kill -0 "$pid" 2>/dev/null; do
printf "\r %-36s %s " "$label" "${spin[$i]}"
i=$(( (i + 1) % 10 ))
sleep 0.2
done
wait "$pid"
local rc=$?
if [ $rc -ne 0 ]; then
printf "\r %-36s ✗ FAILED\n" "$label" | tee -a "$LOG_FILE"
exit $rc
fi
local dur=$(( $(date +%s) - start ))
printf "\r %-36s ✓ %dm%02ds\n" "$label" "$(( dur / 60 ))" "$(( dur % 60 ))" | tee -a "$LOG_FILE"
}
check() {
local label="$1" got="$2" expected="$3"
if [ "$got" -eq "$expected" ]; then
printf " %-20s %s / %s ✓\n" "$label:" "$got" "$expected"
else
printf " %-20s %s / %s ✗ MISMATCH\n" "$label:" "$got" "$expected"
fi
}
# ──────────────────────────────────────────────────────────────────────────────
# ─── DATE ARGUMENT ────────────────────────────────────────────────────────────
if [ "${1:-}" = "today" ]; then
TARGET_DATE=$(date +%Y-%m-%d)
elif [ -n "${1:-}" ]; then
TARGET_DATE="$1"
if ! [[ "$TARGET_DATE" =~ ^[0-9]{4}-[0-9]{2}-[0-9]{2}$ ]]; then
echo "ERROR: Invalid date format. Use YYYY-MM-DD, 'today', or no argument for yesterday."
exit 1
fi
else
TARGET_DATE=$(date -v-1d +%Y-%m-%d)
fi
log "========================================"
log "Screenpipe sync starting for: $TARGET_DATE"
log "========================================"
# ─── PREFLIGHT ────────────────────────────────────────────────────────────────
step "Preflight checks"
if [ ! -f "$DB_SRC" ]; then
log "ERROR: Source DB not found at $DB_SRC"; exit 1
fi
printf " %-20s %s (%s)\n" "Source DB:" "OK" "$(du -sh "$DB_SRC" | cut -f1)"
if [ ! -d "$NAS_MOUNT" ]; then
log "ERROR: NAS not mounted at $NAS_MOUNT"; exit 1
fi
printf " %-20s %s\n" "NAS mount:" "OK $NAS_MOUNT"
# Check if DB already synced for this date
DB_ALREADY_SYNCED=false
if [ -f "$NAS_DB" ]; then
EXISTING=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';" 2>/dev/null || echo "0")
if [ "$EXISTING" -gt "0" ]; then
log "Date $TARGET_DATE already has $EXISTING frames in archive — skipping DB sync"
DB_ALREADY_SYNCED=true
else
printf " %-20s %s (%s)\n" "Archive DB:" "exists" "$(du -sh "$NAS_DB" | cut -f1)"
fi
else
printf " %-20s %s\n" "Archive DB:" "will be created"
fi
# Source data dir for this date
DATA_SRC="$HOME/.screenpipe/data/data/$TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
DATA_SIZE=$(du -sh "$DATA_SRC" | cut -f1)
DATA_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
printf " %-20s %s (%s files, %s)\n" "Data dir:" "OK" "$DATA_FILES" "$DATA_SIZE"
else
printf " %-20s %s\n" "Data dir:" "not found — skipping file copy"
fi
# ─── DB SYNC ──────────────────────────────────────────────────────────────────
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── COUNT SOURCE ROWS ────────────────────────────────────────────────────
step "Counting source rows for $TARGET_DATE"
SRC_FRAMES=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
SRC_ELEMENTS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_UI=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
SRC_OCR=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_MEETINGS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start) = '$TARGET_DATE';")
printf " %-20s %s\n" "frames:" "$SRC_FRAMES"
printf " %-20s %s\n" "elements:" "$SRC_ELEMENTS"
printf " %-20s %s\n" "ui_events:" "$SRC_UI"
printf " %-20s %s\n" "ocr_text:" "$SRC_OCR"
printf " %-20s %s\n" "meetings:" "$SRC_MEETINGS"
if [ "$SRC_FRAMES" -eq "0" ]; then
log "No frames found for $TARGET_DATE — skipping DB sync"
DB_ALREADY_SYNCED=true
fi
fi
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── INIT TABLES ──────────────────────────────────────────────────────────
step "Initialising tables, indexes, FTS"
run_sqlite_heredoc "creating tables" "
ATTACH '$NAS_DB' AS nas;
CREATE TABLE IF NOT EXISTS nas.frames AS SELECT * FROM main.frames WHERE 0;
CREATE TABLE IF NOT EXISTS nas.elements AS SELECT * FROM main.elements WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ui_events AS SELECT * FROM main.ui_events WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ocr_text AS SELECT * FROM main.ocr_text WHERE 0;
CREATE TABLE IF NOT EXISTS nas.video_chunks AS SELECT * FROM main.video_chunks WHERE 0;
CREATE TABLE IF NOT EXISTS nas.meetings AS SELECT * FROM main.meetings WHERE 0;
DETACH nas;
"
run_sqlite_heredoc "creating indexes" "
ATTACH '$NAS_DB' AS nas;
CREATE INDEX IF NOT EXISTS nas.idx_frames_timestamp ON frames(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_frames_app_name ON frames(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_window_name ON frames(window_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_video_chunk_id ON frames(video_chunk_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_id ON elements(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_src_role ON elements(frame_id, source, role) WHERE text IS NOT NULL;
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_timestamp ON ui_events(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_app_name ON ui_events(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_frame_id ON ui_events(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_ocr_text_frame_id ON ocr_text(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_meetings_start ON meetings(meeting_start);
CREATE INDEX IF NOT EXISTS nas.idx_video_chunks_device ON video_chunks(device_name);
DETACH nas;
"
run_sqlite_heredoc "creating FTS tables" "
ATTACH '$NAS_DB' AS nas;
CREATE VIRTUAL TABLE IF NOT EXISTS nas.elements_fts USING fts5(
text, role, frame_id UNINDEXED,
content='elements', content_rowid='id', tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.frames_fts USING fts5(
full_text, app_name, window_name, browser_url, id UNINDEXED,
tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.ui_events_fts USING fts5(
text_content,
app_name,
window_title,
element_name,
content='ui_events',
content_rowid='id',
tokenize='unicode61'
);
DETACH nas;
"
# ─── SYNC DATA ────────────────────────────────────────────────────────────
step "Syncing data for $TARGET_DATE"
run_sqlite_heredoc "video_chunks" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.video_chunks
SELECT * FROM main.video_chunks
WHERE id IN (
SELECT DISTINCT video_chunk_id FROM main.frames
WHERE date(timestamp) = '$TARGET_DATE' AND video_chunk_id IS NOT NULL
);
DETACH nas;
"
run_sqlite_heredoc "frames ($SRC_FRAMES rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.frames
SELECT * FROM main.frames WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ocr_text ($SRC_OCR rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ocr_text
SELECT o.* FROM main.ocr_text o
JOIN main.frames f ON o.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ui_events ($SRC_UI rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ui_events
SELECT * FROM main.ui_events WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "elements ($SRC_ELEMENTS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.elements
SELECT e.* FROM main.elements e
JOIN main.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "meetings ($SRC_MEETINGS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.meetings
SELECT * FROM main.meetings WHERE date(meeting_start) = '$TARGET_DATE';
DETACH nas;
"
# ─── FTS UPDATE ───────────────────────────────────────────────────────────
step "Updating FTS indexes"
run_sqlite_heredoc "elements_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.elements_fts(rowid, text, role)
SELECT e.id, e.text, e.role
FROM nas.elements e
JOIN nas.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE'
AND e.text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "frames_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.frames_fts(rowid, full_text, app_name, window_name, browser_url, id)
SELECT id, full_text, app_name, window_name, browser_url, id
FROM nas.frames
WHERE date(timestamp) = '$TARGET_DATE'
AND full_text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "ui_events_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.ui_events_fts(rowid, text_content, app_name, window_title, element_name)
SELECT id, text_content, app_name, window_title, element_name
FROM nas.ui_events
WHERE date(timestamp) = '$TARGET_DATE'
AND text_content IS NOT NULL;
DETACH nas;
"
# ─── VERIFY DB ────────────────────────────────────────────────────────────
step "Verifying DB"
V_FRAMES=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
V_ELEMENTS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_UI=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
V_OCR=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_MEETINGS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start)= '$TARGET_DATE';")
check "frames" "$V_FRAMES" "$SRC_FRAMES"
check "elements" "$V_ELEMENTS" "$SRC_ELEMENTS"
check "ui_events" "$V_UI" "$SRC_UI"
check "ocr_text" "$V_OCR" "$SRC_OCR"
check "meetings" "$V_MEETINGS" "$SRC_MEETINGS"
fi
# ─── COPY DATA FOLDER ─────────────────────────────────────────────────────────
# Always runs regardless of DB sync status
step "Copying data folder for $TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
mkdir -p "$NAS_DATA/$TARGET_DATE"
RSYNC_START=$(date +%s)
printf " %-36s " "rsync $TARGET_DATE/ → NAS"
rsync -a --ignore-existing \
"$DATA_SRC/" \
"$NAS_DATA/$TARGET_DATE/" \
2>>"$LOG_FILE"
RSYNC_DUR=$(( $(date +%s) - RSYNC_START ))
COPIED_FILES=$(ls "$NAS_DATA/$TARGET_DATE" | wc -l | tr -d ' ')
SRC_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
COPIED_SIZE=$(du -sh "$NAS_DATA/$TARGET_DATE" | cut -f1)
if [ "$COPIED_FILES" -eq "$SRC_FILES" ]; then
printf "\r %-36s ✓ %dm%02ds (%s files, %s)\n" \
"rsync $TARGET_DATE/ → NAS" \
"$(( RSYNC_DUR / 60 ))" "$(( RSYNC_DUR % 60 ))" \
"$COPIED_FILES" "$COPIED_SIZE" | tee -a "$LOG_FILE"
else
printf "\r %-36s ✗ %s / %s files\n" \
"rsync $TARGET_DATE/ → NAS" "$COPIED_FILES" "$SRC_FILES" | tee -a "$LOG_FILE"
fi
else
printf " %-36s %s\n" "rsync $TARGET_DATE/ → NAS" "skipped (no source dir)"
fi
# ─── SUMMARY ──────────────────────────────────────────────────────────────────
TOTAL_ELAPSED=$(( $(date +%s) - SCRIPT_START ))
DB_SIZE=$(du -sh "$NAS_DB" | cut -f1)
echo ""
log "Archive DB size: $DB_SIZE"
log "Total time: $(( TOTAL_ELAPSED / 60 ))m$(( TOTAL_ELAPSED % 60 ))s"
log "Sync complete for $TARGET_DATE"
log "========================================"
lets add ability to jump…, Editor Group 2
Problems (⇧⌘M)
PROBLEMS
Output (⇧⌘U)
OUTPUT
Debug Console (⇧⌘Y)
DEBUG CONSOLE
Terminal (⌃`)
TERMINAL
Ports
PORTS
remote SSH: nas
SSH: nas
screenpipe (Git) - master*, Checkout Branch/Tag...
master*
screenpipe (Git) - Synchronize Changes
No Problems
0
0
No Ports Forwarded
0
Notifications
Signed out
Signed out
Shell Script
Editor Language Status: $(copilot) No inline suggestion available, Inline suggestions
LF
UTF-8
Spaces: 2
Ln 18, Col 36
Screen Reader Optimized
git-commit Lukas Kovalik (23 hours ago)
Lukas Kovalik (23 hours ago)
Info: Downloading VS Code Server...
lets add ability to jump fro,m search to timeline so we can see the exact screen (also based on monitor and time
Session history
New session
Message actions
This session is being continued from a previous conversation that ran out of context. The summary below covers the earlier portion of the conversation.
Summary:
1. Primary Request and Intent:
The user has been developing a Screenpipe dashboard on a Synology NAS. Requests in chronological order:
- **Q1**: Are db.sqlite-shm and db.sqlite-wal used? → Explained WAL mode; cleaned up files.
- **Q2**: AND combination search (term1 + term2 must both appear) → Added "AND also" input field.
- **Q3**: Explain UTC time issue, show date in results, make Only/Skip apps use multiselect → Implemented all three.
- **Q4**: Point sqlite-web at archive.db, add second sqlite-web for db.sqlite → Updated docker-compose.yml, added two services on ports 8767/8768.
- **Q5**: "still the same" (container not picking up config) → Explained `up -d` vs `restart`.
- **Q6**: Add git, push to gitea.com/lakylak/screenpipe.git → Initialized repo, committed, pushed.
- **Q7**: Add Timetable tab with video playback and timeline → Full implementation with click-to-seek, auto-advancing chunks, legend.
- **Q8**: Add scroll/zoom on timeline, jump to specific time, better awareness → Implemented zoom (1×–32×), scrollable timeline, overview minimap, viewport indicator, Jump-to-time input, screenshot thumbnail.
- **Q9 (most recent)**: Make playing speed more than 8×, fix speed not working, show position on timeline → CSS fix done (wider playhead + time label style), remainder in progress.
2. Key Technical Concepts:
- **SQLite WAL mode**: db.sqlite-shm and db.sqlite-wal are WAL companion files; cleaned with `PRAGMA wal_checkpoint(TRUNCATE)`.
- **SQLite FTS5**: Space-separated terms = AND; `MATCH 'term1 term2'` requires both; OR requires explicit `OR` keyword.
- **FastAPI**: Python backend with route handlers in main.py, consumers pattern for DB queries.
- **Docker Compose**: App image is built (code baked in), not volume-mounted → requires `build` + `up -d` for code changes. `restart` doesn't re-read compose file — need `up -d`.
- **sqlite-web**: Read-only flag `-r`, pre-built image, no rebuild needed for config changes.
- **Video playback**: screenpipe records at 0.1–0.3 fps (timelapse). `seek = offset_index / fps` gives seconds into video chunk. ~245 chunks/day across 2 monitors. Videos may be encoded at standard fps making them very short at native speed.
- **Timeline rendering**: App-colored segments built by merging consecutive frames with same `app_name`. Percentage positions: `((ts - dayStart) / span * 100)%`.
- **Zoom/scroll**: Inner div width = `zoom × 100%`, outer wrapper has `overflow-x:auto`. Playhead position stays as % of inner div, which is correct at all zoom levels.
- **Timer-based speed**: For speed > 2×, browser `playbackRate` is unreliable on low-FPS video; correct approach is `setInterval` advancing `video.currentTime` by `speed * interval_ms / 1000` seconds per tick.
- **UTC vs local time**: Backend treats `time_from`/`time_to` as UTC. Frontend converts with `localTimeToUtc()` before sending.
- **Git**: `#recycle/` in .gitignore treated as comment — must escape as `\#recycle/`.
3. Files and Code Sections:
- **`/volume1/screenpipe/app/main.py`**
- Added `/api/apps` endpoint (all unique app names from frames table)
- Added `/api/timetable` endpoint (frames + video URLs + screenshot URLs per device per date)
- `/api/timetable` response structure: `{frames: [{ts, app, win, video_url, seek, screenshot_url}], devices: [...], device: "..."}`
- Key snippet:
```python
@app.get("/api/timetable"
)
async def timetable_data(d: str | None = Query(None), device: str | None = Query(None)):
...
rows = conn.execute(f"""
SELECT f.timestamp AS ts, f.app_name AS app, f.window_name AS win,
f.offset_index, f.snapshot_path,
vc.file_path AS video_path, vc.fps AS video_fps
FROM frames f
JOIN video_chunks vc ON f.video_chunk_id = vc.id
WHERE f.timestamp >= ? AND f.timestamp < ?
AND f.video_chunk_id IS NOT NULL {dev_clause}
ORDER BY f.timestamp
""", params).fetchall()
# seek = offset_index / fps, screenshot_url from _snapshot_to_url(snapshot_path)
```
- **`/volume1/screenpipe/app/docker-compose.yml`**
- Changed sqlite-web service name to `sqlite-web-archive` → port 8767 → `/data/archive.db`
- Added `sqlite-web-live` → port 8768 → `/data/db.sqlite`
- screenpipe-app: port 8766, `DB_PATH: /data/db.sqlite`, `TZ: Europe/Sofia`
- **`/volume1/screenpipe/app/static/index.html`** (single large file with all CSS/HTML/JS)
- **Search enhancements**:
- Added `id="sq2"` "AND also" input between search-row1 and search-row2
- Replaced `sOnlyApps`/`sSkipApps` text inputs with custom multiselect components (`id="msOnlyApps"`, `id="msSkipApps"`)
- Updated `buildUrl()` to combine `q + " " + q2` and convert times with `localTimeToUtc()`
- Added `fmtDateTime(ts)` showing "Apr 20 14:08" format
- Added `localTimeToUtc(timeStr)` converting local HH:MM to UTC
- Multiselect state: `const _msState = { msOnlyApps: new Set(), msSkipApps: new Set() }`
- **Timetable tab** (new page):
- Tab button added between Work Report and AI Summary
- `showPage` array updated: `['activity','search','audio','report','timetable','summary']`
- `refresh()` calls `loadTimetable()` when on timetable page
- **Timetable CSS** (key styles):
```css
.tt-scroll-wrap { overflow-x:auto; ... scrollbar-width:thin; }
.tt-inner { position:relative; height:52px; } /* width set dynamically */
.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
.tt-overview-wrap { position:relative; height:14px; ... cursor:pointer; }
.tt-overview-vp { position:absolute; background:rgba(255,255,255,0.12); border:1px solid rgba(255,255,255,0.4); }
```
- **Timetable JS state variables**:
```javascript
let _ttFrames = [], _ttChunks = [], _ttCurChunk = -1;
let _ttDayStart = 0, _ttDayEnd = 0, _ttZoom = 1;
const _ttAppColors = {};
const TT_ZOOMS = [1, 2, 4, 8, 16, 32];
const TT_TICK_MIN = {1:30, 2:15, 4:10, 8:5, 16:2, 32:1};
```
- **Key JS functions**:
- `loadTimetable()` — fetches `/api/timetable`, groups chunks, renders
- `ttZoom(dir)` — changes zoom level, updates inner width, re-renders ticks, scrolls to playhead
- `ttTimelineClick(e)` — uses `inner.getBoundingClientRect()` (works correctly when scrolled)
- `ttOverviewClick(e)` — scrolls timeline + seeks video
- `ttUpdateOverviewVp()` — updates white viewport rect using `wrap.scrollLeft / inner.scrollWidth`
- `ttJumpToTime()` — parses time input, converts to ms, calls `ttSeekToMs`
- `ttSeekToMs(tsMs)` — binary-searches nearest frame, loads chunk, seeks video
- `_ttScrollToPlayhead(force)` — scrolls wrap so playhead stays at ~35% from left when Follow is checked
- `_ttMovePlayhead(tsMs)` — sets `ph.style.left` as percentage, calls `_ttScrollToPlayhead`
- `_ttShowNow(f)` — updates screenshot thumbnail + now-text div
- `ttSyncBtn()` — syncs play/pause button text
- Keyboard: Space=play/pause, ←/→=±10s, Ctrl+←/→=±30s, +/-=zoom
- **`/volume1/screenpipe/.gitignore`**
```
*.db, *.sqlite, *.sqlite-shm, *.sqlite-wal
data/, \#recycle/,
@eaDir/
app_settings.json, app/.env
__pycache__/, *.pyc, .DS_Store
```
4. Errors and Fixes:
- **`#recycle/` in .gitignore treated as comment**: `#` starts a comment in .gitignore. Fixed by escaping: `\#recycle/`. Also had to `git rm --cached "#recycle/desktop.ini"` to unstage it.
- **`sudo docker compose restart sqlite-web` didn't apply config change**: `restart` reuses existing container config, doesn't re-read compose file. Fix: `sudo docker compose up -d` recreates containers with new config.
- **Orphan container holding port 8767**: Old `screenpipe-sqlweb` container still running after rename. Fix: `sudo docker compose up -d --remove-orphans`.
- **`sudo docker` not accessible from Claude Code**: Interactive sudo required. User runs commands in NAS terminal.
- **`git config --global --add safe.directory /volume1/screenpipe`**: Required due to dubious ownership of NAS volume directory.
5. Problem Solving:
- **Time zone confusion in search**: Backend uses UTC for time_from/time_to; frontend displays local time. Solved by `localTimeToUtc()` converting before API call, and showing date+time in results.
- **sqlite-web not showing April 17–21 data**: Container was pointing at db.sqlite (old config), not archive.db. Fixed by updating docker-compose.yml + `up -d`.
- **Video playback speed**: Browser `video.playbackRate` is capped (~16× max in Chrome) and performs poorly for very low FPS (0.1–0.3 fps) screenpipe videos. Each chunk may only be 1–2 seconds of wall-clock encoded video, playing through instantly. Solution: implement timer-based frame-jumping for speed > 2× (in progress).
- **Playhead visibility**: White 2px line was hard to see. Fixed CSS to 3px width + stronger box-shadow + added `.tt-playhead-time` overlay label style.
6. All User Messages:
- "do we use db.sqlite-shm and db.sqlite-wal for anything right now? [image of search UI] Lets add ability to search with combination... Explain how does the search work time seems to work in utc maybe display date next to the time in results. Make only apps and skip apps multiselect with all available apps."
- "doi I need to restart?"
- "there is sudo docker compose"
- "no still the same"
- "add git. I have created gitea repo https://gitea.com/lakylak/screenpipe.git"
- "just do it" (re: git push)
- "how to make sure data in :8767 are accurate and point to archive.db"
- "I can't see older than 16th there should 21st of april. In http://[IP_ADDRESS]:8766/ I can see all but not on sqlite viewer http://[IP_ADDRESS]:8767"
- "ok lets add additional port to read db.sqlite, I would like to see the structure of official db that reads db.sqlite db and one for archive"
- [docker compose output showing orphan container error]
- "looks great. I want to have ability to scroll left and right and also to jump to specific time. I want to be more aware to what I see. Still it should be fast"
- "lets make the playing speed more than 8x"
- "still playing speed is not showing properly still just in one second. also I want to see in timeline where I am"
7. Pending Tasks:
- **Playback speed fix** (in progress, partially done):
- Add `<div class="tt-playhead-time" id="ttPlayheadTime"></div>` inside `ttInner` HTML
- Implement `_ttStartSpeedTimer(speed)` / `_ttStopSpeedTimer()` / `ttSetSpeed(speed)` functions
- Update speed select `onchange` to call `ttSetSpeed(+this.value)`
- Add 64× option to speed select
- Update `_ttMovePlayhead` to also update `ttPlayheadTime` label
- Update `ttSyncBtn` to start/stop timer based on play state + speed
- Update `ttNextChunk` and `ttSeekToMs` to stop timer (restarted by onplay → ttSyncBtn)
- Rebuild Docker image after changes
8. Current Work:
Working on fixing two issues with the Timetable tab:
1. **Playback speed**: `video.playbackRate` doesn't work well for low-FPS screenpipe videos. The plan is to implement a `setInterval`-based frame-jumper for speed > 2× that advances `video.currentTime` by `speed * intervalMs / 1000` every 100–200ms, bypassing browser playbackRate limits.
2. **Timeline position indicator**: Playhead not prominent enough.
**CSS already changed** in `/volume1/screenpipe/app/static/index.html`:
```css
.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
```
Session was interrupted by summary request after this CSS edit, mid-implementation.
9. Optional Next Step:
Continuing the fix for: "still playing speed is not showing properly still just in one second. also I want to see in timeline where I am"
The next steps in order are:
1. Add `<div class="tt-playhead-time" id="ttPlayheadTime"></div>` to the HTML inside `#ttInner` (after the existing `.tt-playhead` div)
2. Update `_ttMovePlayhead` to set `ttPlayheadTime` position and text content with `fmtTime()`
3. Add timer-based speed system to JS (new functions `_ttStartSpeedTimer`, `_ttStopSpeedTimer`, `ttSetSpeed`)
4. Update speed select's `onchange` to `ttSetSpeed(+this.value)` and add 64× option
5. Update `ttSyncBtn`, `ttNextChunk`, `ttSeekToMs` to integrate with timer
6. Rebuild: `sudo docker compose build screenpipe-app && sudo docker compose up -d screenpipe-app`
If you need specific details from before compaction (like exact code snippets, error messages, or content you generated), read the full transcript at: /home/Adm1n/.claude/projects/-volume1-screenpipe/7777f46e-1e40-4051-b4bb-f43aedd59556.jsonl
Continue the conversation from where it left off without asking the user any further questions. Resume directly — do not acknowledge the summary, do not recap what was happening, do not preface with "I'll continue" or similar. Pick up the last task as if the break never happened.
Thinking
Thinking
Bash
Find timetable speed and playhead related code
IN
grep -n "tt-playhead\|ttPlayhead\|ttInner\|_ttMovePlayhead\|ttSetSpeed\|_ttStartSpeed\|_ttStopSpeed\|ttSyncBtn\|playbackRate\|speed.*select\|spdSel\|ttSpeed" /volume1/screenpipe/app/static/index.html | head -80
Copy code to clipboard
OUT
207:.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
208:.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
580: <div class="tt-inner" id="ttInner" onclick="ttTimelineClick(event)">
583: <div class="tt-playhead" id="ttPlayhead"></div>
598: onpause="ttSyncBtn()"
599: onplay="ttSyncBtn()">
607: <select id="ttSpeed" onchange="document.getElementById('ttVideo').playbackRate=+this.value">
1624: document.getElementById('ttInner').style.width = '100%';
1632: document.getElementById('ttPlayhead').style.display = 'none';
1694: document.getElementById('ttInner').style.width = (_ttZoom * 100) + '%';
1700: const inner = document.getElementById('ttInner');
1710: const inner = document.getElementById('ttInner');
1718: const inner = document.getElementById('ttInner');
1750: _ttMovePlayhead(new Date(best.ts).getTime());
1772: _ttMovePlayhead(new Date(best.ts).getTime());
1791:function _ttMovePlayhead(tsMs) {
1792: const ph = document.getElementById('ttPlayhead');
1802: const ph = document.getElementById('ttPlayhead');
1805: const inner = document.getElementById('ttInner');
1819:function ttSyncBtn() {
Read
index.html
index.html
(lines 576-625)
Read
index.html
index.html
(lines 1786-1865)
Thinking
Thinking
Bash
Find chunk loading and playhead update functions
IN
grep -n "ttUpdatePlayhead\|ttNextChunk\|ttLoadChunk\|_ttCurChunk\|_ttChunks\|_ttFrames" /volume1/screenpipe/app/static/index.html | head -50
Copy code to clipboard
OUT
596: onended="ttNextChunk()"
597: ontimeupdate="ttUpdatePlayhead()"
1569:let _ttFrames = [];
1570:let _ttChunks = [];
1571:let _ttCurChunk = -1;
1607: _ttFrames = data.frames;
1609: for (const f of _ttFrames) {
1613: _ttChunks = [...chunkMap.values()];
1614: _ttCurChunk = -1;
1616: _ttDayStart = new Date(_ttFrames[0].ts).getTime();
1617: _ttDayEnd = new Date(_ttFrames[_ttFrames.length - 1].ts).getTime();
1620: for (const f of _ttFrames) _ttColor(f.app);
1640: for (const f of _ttFrames) {
1736: if (!_ttFrames.length) return;
1737: let best = _ttFrames[0], minDiff = Infinity;
1738: for (const f of _ttFrames) {
1742: const ci = _ttChunks.findIndex(ch => ch[0].video_url === best.video_url);
1744: _ttCurChunk = ci;
1753:function ttNextChunk() {
1754: if (_ttCurChunk + 1 >= _ttChunks.length) return;
1755: _ttCurChunk++;
1757: video.src = _ttChunks[_ttCurChunk][0].video_url;
1762:function ttUpdatePlayhead() {
1764: if (_ttCurChunk < 0 || _ttCurChunk >= _ttChunks.length) return;
1765: const chunk = _ttChunks[_ttCurChunk];
Read
index.html
index.html
(lines 1754-1793)
Thinking
Thinking
Read
index.html
index.html
(lines 1701-1760)
Thinking
Thinking
Read
index.html
index.html
(lines 1561-1640)
Thinking
Thinking
Bash
Find time formatting functions
IN
grep -n "function fmtTime\|function fmtDate\|function _fmt\|localTimeToUtc\|fmtDateTime" /volume1/screenpipe/app/static/index.html | head -20
Copy code to clipboard...
|
Code
|
screenpipe_sync.sh — screenpipe [SSH: nas]
|
NULL
|
|
Explorer (⇧⌘E)
Search (⇧⌘F)
Source Control (⌃⇧ Explorer (⇧⌘E)
Search (⇧⌘F)
Source Control (⌃⇧G) - 2 pending changes
2
Run and Debug (⇧⌘D)
Remote Explorer
Extensions (⇧⌘X) - 2 require update, 1 requires restart
3
Claude Code
Containers
EXPLORER
EXPLORER
Explorer Section: screenpipe [SSH: nas]
Explorer Section: screenpipe [SSH: nas]
SCREENPIPE [SSH: NAS]
#recycle
app
data
pipes
.gitignore
app_settings.json
archive.db
db.sqlite
screenpipe_sync.sh
screenpipe.db
Outline Section
OUTLINE
OUTLINE
Timeline Section
TIMELINE
TIMELINE
app_settings.json, Editor Group 1
screenpipe_sync.sh, Editor Group 1
db.sqlite-shm, preview, Editor Group 1
#!/bin/bash
# screenpipe_sync.sh
# Syncs Screenpipe SQLite data to a NAS archive database (append-only, no deletions).
# Also copies the day's video/frame data folder to the NAS.
#
# Usage:
# ./screenpipe_sync.sh # syncs yesterday (default)
# ./screenpipe_sync.sh 2026-04-15 # syncs a specific date
# ./screenpipe_sync.sh today # syncs today so far
#
# Cron example (runs at 3am daily):
# 0 3 * * * /Users/lukas/.screenpipe/screenpipe_sync.sh >> /Users/lukas/.screenpipe/sync.log 2>&1
set -euo pipefail
# ─── CONFIG ───────────────────────────────────────────────────────────────────
DB_SRC="${SCREENPIPE_DB:-$HOME/.screenpipe/db.sqlite}"
NAS_MOUNT="${NAS_MOUNT:-/Volumes/Test/screenpipe}"
NAS_DB="$NAS_MOUNT/archive.db"
NAS_DATA="$NAS_MOUNT/data"
LOG_FILE="$HOME/.screenpipe/sync.log"
# ──────────────────────────────────────────────────────────────────────────────
# ─── HELPERS ──────────────────────────────────────────────────────────────────
SCRIPT_START=$(date +%s)
log() {
local msg="[$(date '+%Y-%m-%d %H:%M:%S')] $*"
echo "$msg" | tee -a "$LOG_FILE"
}
step() {
local now=$(date +%s)
local elapsed=$(( now - SCRIPT_START ))
local min=$(( elapsed / 60 ))
local sec=$(( elapsed % 60 ))
printf "\n[+%02dm%02ds] ▶ %s\n" "$min" "$sec" "$*" | tee -a "$LOG_FILE"
}
run_sqlite_heredoc() {
local label="$1"
local sql="$2"
local start=$(date +%s)
printf " %-36s " "$label"
sqlite3 "$DB_SRC" <<< "$sql" &
local pid=$!
local spin=[PASSWORD] '⠙' '⠹' '⠸' '⠼' '⠴' '⠦' '⠧' '⠇' '⠏')
local i=0
while kill -0 "$pid" 2>/dev/null; do
printf "\r %-36s %s " "$label" "${spin[$i]}"
i=$(( (i + 1) % 10 ))
sleep 0.2
done
wait "$pid"
local rc=$?
if [ $rc -ne 0 ]; then
printf "\r %-36s ✗ FAILED\n" "$label" | tee -a "$LOG_FILE"
exit $rc
fi
local dur=$(( $(date +%s) - start ))
printf "\r %-36s ✓ %dm%02ds\n" "$label" "$(( dur / 60 ))" "$(( dur % 60 ))" | tee -a "$LOG_FILE"
}
check() {
local label="$1" got="$2" expected="$3"
if [ "$got" -eq "$expected" ]; then
printf " %-20s %s / %s ✓\n" "$label:" "$got" "$expected"
else
printf " %-20s %s / %s ✗ MISMATCH\n" "$label:" "$got" "$expected"
fi
}
# ──────────────────────────────────────────────────────────────────────────────
# ─── DATE ARGUMENT ────────────────────────────────────────────────────────────
if [ "${1:-}" = "today" ]; then
TARGET_DATE=$(date +%Y-%m-%d)
elif [ -n "${1:-}" ]; then
TARGET_DATE="$1"
if ! [[ "$TARGET_DATE" =~ ^[0-9]{4}-[0-9]{2}-[0-9]{2}$ ]]; then
echo "ERROR: Invalid date format. Use YYYY-MM-DD, 'today', or no argument for yesterday."
exit 1
fi
else
TARGET_DATE=$(date -v-1d +%Y-%m-%d)
fi
log "========================================"
log "Screenpipe sync starting for: $TARGET_DATE"
log "========================================"
# ─── PREFLIGHT ────────────────────────────────────────────────────────────────
step "Preflight checks"
if [ ! -f "$DB_SRC" ]; then
log "ERROR: Source DB not found at $DB_SRC"; exit 1
fi
printf " %-20s %s (%s)\n" "Source DB:" "OK" "$(du -sh "$DB_SRC" | cut -f1)"
if [ ! -d "$NAS_MOUNT" ]; then
log "ERROR: NAS not mounted at $NAS_MOUNT"; exit 1
fi
printf " %-20s %s\n" "NAS mount:" "OK $NAS_MOUNT"
# Check if DB already synced for this date
DB_ALREADY_SYNCED=false
if [ -f "$NAS_DB" ]; then
EXISTING=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';" 2>/dev/null || echo "0")
if [ "$EXISTING" -gt "0" ]; then
log "Date $TARGET_DATE already has $EXISTING frames in archive — skipping DB sync"
DB_ALREADY_SYNCED=true
else
printf " %-20s %s (%s)\n" "Archive DB:" "exists" "$(du -sh "$NAS_DB" | cut -f1)"
fi
else
printf " %-20s %s\n" "Archive DB:" "will be created"
fi
# Source data dir for this date
DATA_SRC="$HOME/.screenpipe/data/data/$TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
DATA_SIZE=$(du -sh "$DATA_SRC" | cut -f1)
DATA_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
printf " %-20s %s (%s files, %s)\n" "Data dir:" "OK" "$DATA_FILES" "$DATA_SIZE"
else
printf " %-20s %s\n" "Data dir:" "not found — skipping file copy"
fi
# ─── DB SYNC ──────────────────────────────────────────────────────────────────
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── COUNT SOURCE ROWS ────────────────────────────────────────────────────
step "Counting source rows for $TARGET_DATE"
SRC_FRAMES=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
SRC_ELEMENTS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_UI=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
SRC_OCR=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_MEETINGS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start) = '$TARGET_DATE';")
printf " %-20s %s\n" "frames:" "$SRC_FRAMES"
printf " %-20s %s\n" "elements:" "$SRC_ELEMENTS"
printf " %-20s %s\n" "ui_events:" "$SRC_UI"
printf " %-20s %s\n" "ocr_text:" "$SRC_OCR"
printf " %-20s %s\n" "meetings:" "$SRC_MEETINGS"
if [ "$SRC_FRAMES" -eq "0" ]; then
log "No frames found for $TARGET_DATE — skipping DB sync"
DB_ALREADY_SYNCED=true
fi
fi
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── INIT TABLES ──────────────────────────────────────────────────────────
step "Initialising tables, indexes, FTS"
run_sqlite_heredoc "creating tables" "
ATTACH '$NAS_DB' AS nas;
CREATE TABLE IF NOT EXISTS nas.frames AS SELECT * FROM main.frames WHERE 0;
CREATE TABLE IF NOT EXISTS nas.elements AS SELECT * FROM main.elements WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ui_events AS SELECT * FROM main.ui_events WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ocr_text AS SELECT * FROM main.ocr_text WHERE 0;
CREATE TABLE IF NOT EXISTS nas.video_chunks AS SELECT * FROM main.video_chunks WHERE 0;
CREATE TABLE IF NOT EXISTS nas.meetings AS SELECT * FROM main.meetings WHERE 0;
DETACH nas;
"
run_sqlite_heredoc "creating indexes" "
ATTACH '$NAS_DB' AS nas;
CREATE INDEX IF NOT EXISTS nas.idx_frames_timestamp ON frames(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_frames_app_name ON frames(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_window_name ON frames(window_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_video_chunk_id ON frames(video_chunk_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_id ON elements(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_src_role ON elements(frame_id, source, role) WHERE text IS NOT NULL;
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_timestamp ON ui_events(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_app_name ON ui_events(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_frame_id ON ui_events(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_ocr_text_frame_id ON ocr_text(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_meetings_start ON meetings(meeting_start);
CREATE INDEX IF NOT EXISTS nas.idx_video_chunks_device ON video_chunks(device_name);
DETACH nas;
"
run_sqlite_heredoc "creating FTS tables" "
ATTACH '$NAS_DB' AS nas;
CREATE VIRTUAL TABLE IF NOT EXISTS nas.elements_fts USING fts5(
text, role, frame_id UNINDEXED,
content='elements', content_rowid='id', tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.frames_fts USING fts5(
full_text, app_name, window_name, browser_url, id UNINDEXED,
tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.ui_events_fts USING fts5(
text_content,
app_name,
window_title,
element_name,
content='ui_events',
content_rowid='id',
tokenize='unicode61'
);
DETACH nas;
"
# ─── SYNC DATA ────────────────────────────────────────────────────────────
step "Syncing data for $TARGET_DATE"
run_sqlite_heredoc "video_chunks" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.video_chunks
SELECT * FROM main.video_chunks
WHERE id IN (
SELECT DISTINCT video_chunk_id FROM main.frames
WHERE date(timestamp) = '$TARGET_DATE' AND video_chunk_id IS NOT NULL
);
DETACH nas;
"
run_sqlite_heredoc "frames ($SRC_FRAMES rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.frames
SELECT * FROM main.frames WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ocr_text ($SRC_OCR rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ocr_text
SELECT o.* FROM main.ocr_text o
JOIN main.frames f ON o.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ui_events ($SRC_UI rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ui_events
SELECT * FROM main.ui_events WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "elements ($SRC_ELEMENTS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.elements
SELECT e.* FROM main.elements e
JOIN main.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "meetings ($SRC_MEETINGS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.meetings
SELECT * FROM main.meetings WHERE date(meeting_start) = '$TARGET_DATE';
DETACH nas;
"
# ─── FTS UPDATE ───────────────────────────────────────────────────────────
step "Updating FTS indexes"
run_sqlite_heredoc "elements_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.elements_fts(rowid, text, role)
SELECT e.id, e.text, e.role
FROM nas.elements e
JOIN nas.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE'
AND e.text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "frames_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.frames_fts(rowid, full_text, app_name, window_name, browser_url, id)
SELECT id, full_text, app_name, window_name, browser_url, id
FROM nas.frames
WHERE date(timestamp) = '$TARGET_DATE'
AND full_text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "ui_events_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.ui_events_fts(rowid, text_content, app_name, window_title, element_name)
SELECT id, text_content, app_name, window_title, element_name
FROM nas.ui_events
WHERE date(timestamp) = '$TARGET_DATE'
AND text_content IS NOT NULL;
DETACH nas;
"
# ─── VERIFY DB ────────────────────────────────────────────────────────────
step "Verifying DB"
V_FRAMES=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
V_ELEMENTS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_UI=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
V_OCR=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_MEETINGS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start)= '$TARGET_DATE';")
check "frames" "$V_FRAMES" "$SRC_FRAMES"
check "elements" "$V_ELEMENTS" "$SRC_ELEMENTS"
check "ui_events" "$V_UI" "$SRC_UI"
check "ocr_text" "$V_OCR" "$SRC_OCR"
check "meetings" "$V_MEETINGS" "$SRC_MEETINGS"
fi
# ─── COPY DATA FOLDER ─────────────────────────────────────────────────────────
# Always runs regardless of DB sync status
step "Copying data folder for $TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
mkdir -p "$NAS_DATA/$TARGET_DATE"
RSYNC_START=$(date +%s)
printf " %-36s " "rsync $TARGET_DATE/ → NAS"
rsync -a --ignore-existing \
"$DATA_SRC/" \
"$NAS_DATA/$TARGET_DATE/" \
2>>"$LOG_FILE"
RSYNC_DUR=$(( $(date +%s) - RSYNC_START ))
COPIED_FILES=$(ls "$NAS_DATA/$TARGET_DATE" | wc -l | tr -d ' ')
SRC_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
COPIED_SIZE=$(du -sh "$NAS_DATA/$TARGET_DATE" | cut -f1)
if [ "$COPIED_FILES" -eq "$SRC_FILES" ]; then
printf "\r %-36s ✓ %dm%02ds (%s files, %s)\n" \
"rsync $TARGET_DATE/ → NAS" \
"$(( RSYNC_DUR / 60 ))" "$(( RSYNC_DUR % 60 ))" \
"$COPIED_FILES" "$COPIED_SIZE" | tee -a "$LOG_FILE"
else
printf "\r %-36s ✗ %s / %s files\n" \
"rsync $TARGET_DATE/ → NAS" "$COPIED_FILES" "$SRC_FILES" | tee -a "$LOG_FILE"
fi
else
printf " %-36s %s\n" "rsync $TARGET_DATE/ → NAS" "skipped (no source dir)"
fi
# ─── SUMMARY ──────────────────────────────────────────────────────────────────
TOTAL_ELAPSED=$(( $(date +%s) - SCRIPT_START ))
DB_SIZE=$(du -sh "$NAS_DB" | cut -f1)
echo ""
log "Archive DB size: $DB_SIZE"
log "Total time: $(( TOTAL_ELAPSED / 60 ))m$(( TOTAL_ELAPSED % 60 ))s"
log "Sync complete for $TARGET_DATE"
log "========================================"
#!/bin/bash
# screenpipe_sync.sh
# Syncs Screenpipe SQLite data to a NAS archive database (append-only, no deletions).
# Also copies the day's video/frame data folder to the NAS.
#
# Usage:
# ./screenpipe_sync.sh # syncs yesterday (default)
# ./screenpipe_sync.sh 2026-04-15 # syncs a specific date
# ./screenpipe_sync.sh today # syncs today so far
#
# Cron example (runs at 3am daily):
# 0 3 * * * /Users/lukas/.screenpipe/screenpipe_sync.sh >> /Users/lukas/.screenpipe/sync.log 2>&1
set -euo pipefail
# ─── CONFIG ───────────────────────────────────────────────────────────────────
DB_SRC="${SCREENPIPE_DB:-$HOME/.screenpipe/db.sqlite}"
NAS_MOUNT="${NAS_MOUNT:-/Volumes/Test/screenpipe}"
NAS_DB="$NAS_MOUNT/archive.db"
NAS_DATA="$NAS_MOUNT/data"
LOG_FILE="$HOME/.screenpipe/sync.log"
# ──────────────────────────────────────────────────────────────────────────────
# ─── HELPERS ──────────────────────────────────────────────────────────────────
SCRIPT_START=$(date +%s)
log() {
local msg="[$(date '+%Y-%m-%d %H:%M:%S')] $*"
echo "$msg" | tee -a "$LOG_FILE"
}
step() {
local now=$(date +%s)
local elapsed=$(( now - SCRIPT_START ))
local min=$(( elapsed / 60 ))
local sec=$(( elapsed % 60 ))
printf "\n[+%02dm%02ds] ▶ %s\n" "$min" "$sec" "$*" | tee -a "$LOG_FILE"
}
run_sqlite_heredoc() {
local label="$1"
local sql="$2"
local start=$(date +%s)
printf " %-36s " "$label"
sqlite3 "$DB_SRC" <<< "$sql" &
local pid=$!
local spin=[PASSWORD] '⠙' '⠹' '⠸' '⠼' '⠴' '⠦' '⠧' '⠇' '⠏')
local i=0
while kill -0 "$pid" 2>/dev/null; do
printf "\r %-36s %s " "$label" "${spin[$i]}"
i=$(( (i + 1) % 10 ))
sleep 0.2
done
wait "$pid"
local rc=$?
if [ $rc -ne 0 ]; then
printf "\r %-36s ✗ FAILED\n" "$label" | tee -a "$LOG_FILE"
exit $rc
fi
local dur=$(( $(date +%s) - start ))
printf "\r %-36s ✓ %dm%02ds\n" "$label" "$(( dur / 60 ))" "$(( dur % 60 ))" | tee -a "$LOG_FILE"
}
check() {
local label="$1" got="$2" expected="$3"
if [ "$got" -eq "$expected" ]; then
printf " %-20s %s / %s ✓\n" "$label:" "$got" "$expected"
else
printf " %-20s %s / %s ✗ MISMATCH\n" "$label:" "$got" "$expected"
fi
}
# ──────────────────────────────────────────────────────────────────────────────
# ─── DATE ARGUMENT ────────────────────────────────────────────────────────────
if [ "${1:-}" = "today" ]; then
TARGET_DATE=$(date +%Y-%m-%d)
elif [ -n "${1:-}" ]; then
TARGET_DATE="$1"
if ! [[ "$TARGET_DATE" =~ ^[0-9]{4}-[0-9]{2}-[0-9]{2}$ ]]; then
echo "ERROR: Invalid date format. Use YYYY-MM-DD, 'today', or no argument for yesterday."
exit 1
fi
else
TARGET_DATE=$(date -v-1d +%Y-%m-%d)
fi
log "========================================"
log "Screenpipe sync starting for: $TARGET_DATE"
log "========================================"
# ─── PREFLIGHT ────────────────────────────────────────────────────────────────
step "Preflight checks"
if [ ! -f "$DB_SRC" ]; then
log "ERROR: Source DB not found at $DB_SRC"; exit 1
fi
printf " %-20s %s (%s)\n" "Source DB:" "OK" "$(du -sh "$DB_SRC" | cut -f1)"
if [ ! -d "$NAS_MOUNT" ]; then
log "ERROR: NAS not mounted at $NAS_MOUNT"; exit 1
fi
printf " %-20s %s\n" "NAS mount:" "OK $NAS_MOUNT"
# Check if DB already synced for this date
DB_ALREADY_SYNCED=false
if [ -f "$NAS_DB" ]; then
EXISTING=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';" 2>/dev/null || echo "0")
if [ "$EXISTING" -gt "0" ]; then
log "Date $TARGET_DATE already has $EXISTING frames in archive — skipping DB sync"
DB_ALREADY_SYNCED=true
else
printf " %-20s %s (%s)\n" "Archive DB:" "exists" "$(du -sh "$NAS_DB" | cut -f1)"
fi
else
printf " %-20s %s\n" "Archive DB:" "will be created"
fi
# Source data dir for this date
DATA_SRC="$HOME/.screenpipe/data/data/$TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
DATA_SIZE=$(du -sh "$DATA_SRC" | cut -f1)
DATA_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
printf " %-20s %s (%s files, %s)\n" "Data dir:" "OK" "$DATA_FILES" "$DATA_SIZE"
else
printf " %-20s %s\n" "Data dir:" "not found — skipping file copy"
fi
# ─── DB SYNC ──────────────────────────────────────────────────────────────────
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── COUNT SOURCE ROWS ────────────────────────────────────────────────────
step "Counting source rows for $TARGET_DATE"
SRC_FRAMES=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
SRC_ELEMENTS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_UI=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
SRC_OCR=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_MEETINGS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start) = '$TARGET_DATE';")
printf " %-20s %s\n" "frames:" "$SRC_FRAMES"
printf " %-20s %s\n" "elements:" "$SRC_ELEMENTS"
printf " %-20s %s\n" "ui_events:" "$SRC_UI"
printf " %-20s %s\n" "ocr_text:" "$SRC_OCR"
printf " %-20s %s\n" "meetings:" "$SRC_MEETINGS"
if [ "$SRC_FRAMES" -eq "0" ]; then
log "No frames found for $TARGET_DATE — skipping DB sync"
DB_ALREADY_SYNCED=true
fi
fi
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── INIT TABLES ──────────────────────────────────────────────────────────
step "Initialising tables, indexes, FTS"
run_sqlite_heredoc "creating tables" "
ATTACH '$NAS_DB' AS nas;
CREATE TABLE IF NOT EXISTS nas.frames AS SELECT * FROM main.frames WHERE 0;
CREATE TABLE IF NOT EXISTS nas.elements AS SELECT * FROM main.elements WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ui_events AS SELECT * FROM main.ui_events WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ocr_text AS SELECT * FROM main.ocr_text WHERE 0;
CREATE TABLE IF NOT EXISTS nas.video_chunks AS SELECT * FROM main.video_chunks WHERE 0;
CREATE TABLE IF NOT EXISTS nas.meetings AS SELECT * FROM main.meetings WHERE 0;
DETACH nas;
"
run_sqlite_heredoc "creating indexes" "
ATTACH '$NAS_DB' AS nas;
CREATE INDEX IF NOT EXISTS nas.idx_frames_timestamp ON frames(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_frames_app_name ON frames(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_window_name ON frames(window_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_video_chunk_id ON frames(video_chunk_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_id ON elements(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_src_role ON elements(frame_id, source, role) WHERE text IS NOT NULL;
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_timestamp ON ui_events(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_app_name ON ui_events(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_frame_id ON ui_events(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_ocr_text_frame_id ON ocr_text(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_meetings_start ON meetings(meeting_start);
CREATE INDEX IF NOT EXISTS nas.idx_video_chunks_device ON video_chunks(device_name);
DETACH nas;
"
run_sqlite_heredoc "creating FTS tables" "
ATTACH '$NAS_DB' AS nas;
CREATE VIRTUAL TABLE IF NOT EXISTS nas.elements_fts USING fts5(
text, role, frame_id UNINDEXED,
content='elements', content_rowid='id', tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.frames_fts USING fts5(
full_text, app_name, window_name, browser_url, id UNINDEXED,
tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.ui_events_fts USING fts5(
text_content,
app_name,
window_title,
element_name,
content='ui_events',
content_rowid='id',
tokenize='unicode61'
);
DETACH nas;
"
# ─── SYNC DATA ────────────────────────────────────────────────────────────
step "Syncing data for $TARGET_DATE"
run_sqlite_heredoc "video_chunks" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.video_chunks
SELECT * FROM main.video_chunks
WHERE id IN (
SELECT DISTINCT video_chunk_id FROM main.frames
WHERE date(timestamp) = '$TARGET_DATE' AND video_chunk_id IS NOT NULL
);
DETACH nas;
"
run_sqlite_heredoc "frames ($SRC_FRAMES rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.frames
SELECT * FROM main.frames WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ocr_text ($SRC_OCR rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ocr_text
SELECT o.* FROM main.ocr_text o
JOIN main.frames f ON o.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ui_events ($SRC_UI rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ui_events
SELECT * FROM main.ui_events WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "elements ($SRC_ELEMENTS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.elements
SELECT e.* FROM main.elements e
JOIN main.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "meetings ($SRC_MEETINGS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.meetings
SELECT * FROM main.meetings WHERE date(meeting_start) = '$TARGET_DATE';
DETACH nas;
"
# ─── FTS UPDATE ───────────────────────────────────────────────────────────
step "Updating FTS indexes"
run_sqlite_heredoc "elements_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.elements_fts(rowid, text, role)
SELECT e.id, e.text, e.role
FROM nas.elements e
JOIN nas.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE'
AND e.text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "frames_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.frames_fts(rowid, full_text, app_name, window_name, browser_url, id)
SELECT id, full_text, app_name, window_name, browser_url, id
FROM nas.frames
WHERE date(timestamp) = '$TARGET_DATE'
AND full_text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "ui_events_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.ui_events_fts(rowid, text_content, app_name, window_title, element_name)
SELECT id, text_content, app_name, window_title, element_name
FROM nas.ui_events
WHERE date(timestamp) = '$TARGET_DATE'
AND text_content IS NOT NULL;
DETACH nas;
"
# ─── VERIFY DB ────────────────────────────────────────────────────────────
step "Verifying DB"
V_FRAMES=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
V_ELEMENTS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_UI=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
V_OCR=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_MEETINGS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start)= '$TARGET_DATE';")
check "frames" "$V_FRAMES" "$SRC_FRAMES"
check "elements" "$V_ELEMENTS" "$SRC_ELEMENTS"
check "ui_events" "$V_UI" "$SRC_UI"
check "ocr_text" "$V_OCR" "$SRC_OCR"
check "meetings" "$V_MEETINGS" "$SRC_MEETINGS"
fi
# ─── COPY DATA FOLDER ─────────────────────────────────────────────────────────
# Always runs regardless of DB sync status
step "Copying data folder for $TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
mkdir -p "$NAS_DATA/$TARGET_DATE"
RSYNC_START=$(date +%s)
printf " %-36s " "rsync $TARGET_DATE/ → NAS"
rsync -a --ignore-existing \
"$DATA_SRC/" \
"$NAS_DATA/$TARGET_DATE/" \
2>>"$LOG_FILE"
RSYNC_DUR=$(( $(date +%s) - RSYNC_START ))
COPIED_FILES=$(ls "$NAS_DATA/$TARGET_DATE" | wc -l | tr -d ' ')
SRC_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
COPIED_SIZE=$(du -sh "$NAS_DATA/$TARGET_DATE" | cut -f1)
if [ "$COPIED_FILES" -eq "$SRC_FILES" ]; then
printf "\r %-36s ✓ %dm%02ds (%s files, %s)\n" \
"rsync $TARGET_DATE/ → NAS" \
"$(( RSYNC_DUR / 60 ))" "$(( RSYNC_DUR % 60 ))" \
"$COPIED_FILES" "$COPIED_SIZE" | tee -a "$LOG_FILE"
else
printf "\r %-36s ✗ %s / %s files\n" \
"rsync $TARGET_DATE/ → NAS" "$COPIED_FILES" "$SRC_FILES" | tee -a "$LOG_FILE"
fi
else
printf " %-36s %s\n" "rsync $TARGET_DATE/ → NAS" "skipped (no source dir)"
fi
# ─── SUMMARY ──────────────────────────────────────────────────────────────────
TOTAL_ELAPSED=$(( $(date +%s) - SCRIPT_START ))
DB_SIZE=$(du -sh "$NAS_DB" | cut -f1)
echo ""
log "Archive DB size: $DB_SIZE"
log "Total time: $(( TOTAL_ELAPSED / 60 ))m$(( TOTAL_ELAPSED % 60 ))s"
log "Sync complete for $TARGET_DATE"
log "========================================"
lets add ability to jump…, Editor Group 2
Problems (⇧⌘M)
PROBLEMS
Output (⇧⌘U)
OUTPUT
Debug Console (⇧⌘Y)
DEBUG CONSOLE
Terminal (⌃`)
TERMINAL
Ports
PORTS
remote SSH: nas
SSH: nas
screenpipe (Git) - master*, Checkout Branch/Tag...
master*
screenpipe (Git) - Synchronize Changes
No Problems
0
0
No Ports Forwarded
0
Notifications
Signed out
Signed out
Shell Script
Editor Language Status: $(copilot) No inline suggestion available, Inline suggestions
LF
UTF-8
Spaces: 2
Ln 18, Col 36
Screen Reader Optimized
git-commit Lukas Kovalik (23 hours ago)
Lukas Kovalik (23 hours ago)
Info: Downloading VS Code Server...
lets add ability to jump fro,m search to timeline so we can see the exact screen (also based on monitor and time
Session history
New session
Message actions
This session is being continued from a previous conversation that ran out of context. The summary below covers the earlier portion of the conversation.
Summary:
1. Primary Request and Intent:
The user has been developing a Screenpipe dashboard on a Synology NAS. Requests in chronological order:
- **Q1**: Are db.sqlite-shm and db.sqlite-wal used? → Explained WAL mode; cleaned up files.
- **Q2**: AND combination search (term1 + term2 must both appear) → Added "AND also" input field.
- **Q3**: Explain UTC time issue, show date in results, make Only/Skip apps use multiselect → Implemented all three.
- **Q4**: Point sqlite-web at archive.db, add second sqlite-web for db.sqlite → Updated docker-compose.yml, added two services on ports 8767/8768.
- **Q5**: "still the same" (container not picking up config) → Explained `up -d` vs `restart`.
- **Q6**: Add git, push to gitea.com/lakylak/screenpipe.git → Initialized repo, committed, pushed.
- **Q7**: Add Timetable tab with video playback and timeline → Full implementation with click-to-seek, auto-advancing chunks, legend.
- **Q8**: Add scroll/zoom on timeline, jump to specific time, better awareness → Implemented zoom (1×–32×), scrollable timeline, overview minimap, viewport indicator, Jump-to-time input, screenshot thumbnail.
- **Q9 (most recent)**: Make playing speed more than 8×, fix speed not working, show position on timeline → CSS fix done (wider playhead + time label style), remainder in progress.
2. Key Technical Concepts:
- **SQLite WAL mode**: db.sqlite-shm and db.sqlite-wal are WAL companion files; cleaned with `PRAGMA wal_checkpoint(TRUNCATE)`.
- **SQLite FTS5**: Space-separated terms = AND; `MATCH 'term1 term2'` requires both; OR requires explicit `OR` keyword.
- **FastAPI**: Python backend with route handlers in main.py, consumers pattern for DB queries.
- **Docker Compose**: App image is built (code baked in), not volume-mounted → requires `build` + `up -d` for code changes. `restart` doesn't re-read compose file — need `up -d`.
- **sqlite-web**: Read-only flag `-r`, pre-built image, no rebuild needed for config changes.
- **Video playback**: screenpipe records at 0.1–0.3 fps (timelapse). `seek = offset_index / fps` gives seconds into video chunk. ~245 chunks/day across 2 monitors. Videos may be encoded at standard fps making them very short at native speed.
- **Timeline rendering**: App-colored segments built by merging consecutive frames with same `app_name`. Percentage positions: `((ts - dayStart) / span * 100)%`.
- **Zoom/scroll**: Inner div width = `zoom × 100%`, outer wrapper has `overflow-x:auto`. Playhead position stays as % of inner div, which is correct at all zoom levels.
- **Timer-based speed**: For speed > 2×, browser `playbackRate` is unreliable on low-FPS video; correct approach is `setInterval` advancing `video.currentTime` by `speed * interval_ms / 1000` seconds per tick.
- **UTC vs local time**: Backend treats `time_from`/`time_to` as UTC. Frontend converts with `localTimeToUtc()` before sending.
- **Git**: `#recycle/` in .gitignore treated as comment — must escape as `\#recycle/`.
3. Files and Code Sections:
- **`/volume1/screenpipe/app/main.py`**
- Added `/api/apps` endpoint (all unique app names from frames table)
- Added `/api/timetable` endpoint (frames + video URLs + screenshot URLs per device per date)
- `/api/timetable` response structure: `{frames: [{ts, app, win, video_url, seek, screenshot_url}], devices: [...], device: "..."}`
- Key snippet:
```python
@app.get("/api/timetable"
)
async def timetable_data(d: str | None = Query(None), device: str | None = Query(None)):
...
rows = conn.execute(f"""
SELECT f.timestamp AS ts, f.app_name AS app, f.window_name AS win,
f.offset_index, f.snapshot_path,
vc.file_path AS video_path, vc.fps AS video_fps
FROM frames f
JOIN video_chunks vc ON f.video_chunk_id = vc.id
WHERE f.timestamp >= ? AND f.timestamp < ?
AND f.video_chunk_id IS NOT NULL {dev_clause}
ORDER BY f.timestamp
""", params).fetchall()
# seek = offset_index / fps, screenshot_url from _snapshot_to_url(snapshot_path)
```
- **`/volume1/screenpipe/app/docker-compose.yml`**
- Changed sqlite-web service name to `sqlite-web-archive` → port 8767 → `/data/archive.db`
- Added `sqlite-web-live` → port 8768 → `/data/db.sqlite`
- screenpipe-app: port 8766, `DB_PATH: /data/db.sqlite`, `TZ: Europe/Sofia`
- **`/volume1/screenpipe/app/static/index.html`** (single large file with all CSS/HTML/JS)
- **Search enhancements**:
- Added `id="sq2"` "AND also" input between search-row1 and search-row2
- Replaced `sOnlyApps`/`sSkipApps` text inputs with custom multiselect components (`id="msOnlyApps"`, `id="msSkipApps"`)
- Updated `buildUrl()` to combine `q + " " + q2` and convert times with `localTimeToUtc()`
- Added `fmtDateTime(ts)` showing "Apr 20 14:08" format
- Added `localTimeToUtc(timeStr)` converting local HH:MM to UTC
- Multiselect state: `const _msState = { msOnlyApps: new Set(), msSkipApps: new Set() }`
- **Timetable tab** (new page):
- Tab button added between Work Report and AI Summary
- `showPage` array updated: `['activity','search','audio','report','timetable','summary']`
- `refresh()` calls `loadTimetable()` when on timetable page
- **Timetable CSS** (key styles):
```css
.tt-scroll-wrap { overflow-x:auto; ... scrollbar-width:thin; }
.tt-inner { position:relative; height:52px; } /* width set dynamically */
.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
.tt-overview-wrap { position:relative; height:14px; ... cursor:pointer; }
.tt-overview-vp { position:absolute; background:rgba(255,255,255,0.12); border:1px solid rgba(255,255,255,0.4); }
```
- **Timetable JS state variables**:
```javascript
let _ttFrames = [], _ttChunks = [], _ttCurChunk = -1;
let _ttDayStart = 0, _ttDayEnd = 0, _ttZoom = 1;
const _ttAppColors = {};
const TT_ZOOMS = [1, 2, 4, 8, 16, 32];
const TT_TICK_MIN = {1:30, 2:15, 4:10, 8:5, 16:2, 32:1};
```
- **Key JS functions**:
- `loadTimetable()` — fetches `/api/timetable`, groups chunks, renders
- `ttZoom(dir)` — changes zoom level, updates inner width, re-renders ticks, scrolls to playhead
- `ttTimelineClick(e)` — uses `inner.getBoundingClientRect()` (works correctly when scrolled)
- `ttOverviewClick(e)` — scrolls timeline + seeks video
- `ttUpdateOverviewVp()` — updates white viewport rect using `wrap.scrollLeft / inner.scrollWidth`
- `ttJumpToTime()` — parses time input, converts to ms, calls `ttSeekToMs`
- `ttSeekToMs(tsMs)` — binary-searches nearest frame, loads chunk, seeks video
- `_ttScrollToPlayhead(force)` — scrolls wrap so playhead stays at ~35% from left when Follow is checked
- `_ttMovePlayhead(tsMs)` — sets `ph.style.left` as percentage, calls `_ttScrollToPlayhead`
- `_ttShowNow(f)` — updates screenshot thumbnail + now-text div
- `ttSyncBtn()` — syncs play/pause button text
- Keyboard: Space=play/pause, ←/→=±10s, Ctrl+←/→=±30s, +/-=zoom
- **`/volume1/screenpipe/.gitignore`**
```
*.db, *.sqlite, *.sqlite-shm, *.sqlite-wal
data/, \#recycle/,
@eaDir/
app_settings.json, app/.env
__pycache__/, *.pyc, .DS_Store
```
4. Errors and Fixes:
- **`#recycle/` in .gitignore treated as comment**: `#` starts a comment in .gitignore. Fixed by escaping: `\#recycle/`. Also had to `git rm --cached "#recycle/desktop.ini"` to unstage it.
- **`sudo docker compose restart sqlite-web` didn't apply config change**: `restart` reuses existing container config, doesn't re-read compose file. Fix: `sudo docker compose up -d` recreates containers with new config.
- **Orphan container holding port 8767**: Old `screenpipe-sqlweb` container still running after rename. Fix: `sudo docker compose up -d --remove-orphans`.
- **`sudo docker` not accessible from Claude Code**: Interactive sudo required. User runs commands in NAS terminal.
- **`git config --global --add safe.directory /volume1/screenpipe`**: Required due to dubious ownership of NAS volume directory.
5. Problem Solving:
- **Time zone confusion in search**: Backend uses UTC for time_from/time_to; frontend displays local time. Solved by `localTimeToUtc()` converting before API call, and showing date+time in results.
- **sqlite-web not showing April 17–21 data**: Container was pointing at db.sqlite (old config), not archive.db. Fixed by updating docker-compose.yml + `up -d`.
- **Video playback speed**: Browser `video.playbackRate` is capped (~16× max in Chrome) and performs poorly for very low FPS (0.1–0.3 fps) screenpipe videos. Each chunk may only be 1–2 seconds of wall-clock encoded video, playing through instantly. Solution: implement timer-based frame-jumping for speed > 2× (in progress).
- **Playhead visibility**: White 2px line was hard to see. Fixed CSS to 3px width + stronger box-shadow + added `.tt-playhead-time` overlay label style.
6. All User Messages:
- "do we use db.sqlite-shm and db.sqlite-wal for anything right now? [image of search UI] Lets add ability to search with combination... Explain how does the search work time seems to work in utc maybe display date next to the time in results. Make only apps and skip apps multiselect with all available apps."
- "doi I need to restart?"
- "there is sudo docker compose"
- "no still the same"
- "add git. I have created gitea repo https://gitea.com/lakylak/screenpipe.git"
- "just do it" (re: git push)
- "how to make sure data in :8767 are accurate and point to archive.db"
- "I can't see older than 16th there should 21st of april. In http://[IP_ADDRESS]:8766/ I can see all but not on sqlite viewer http://[IP_ADDRESS]:8767"
- "ok lets add additional port to read db.sqlite, I would like to see the structure of official db that reads db.sqlite db and one for archive"
- [docker compose output showing orphan container error]
- "looks great. I want to have ability to scroll left and right and also to jump to specific time. I want to be more aware to what I see. Still it should be fast"
- "lets make the playing speed more than 8x"
- "still playing speed is not showing properly still just in one second. also I want to see in timeline where I am"
7. Pending Tasks:
- **Playback speed fix** (in progress, partially done):
- Add `<div class="tt-playhead-time" id="ttPlayheadTime"></div>` inside `ttInner` HTML
- Implement `_ttStartSpeedTimer(speed)` / `_ttStopSpeedTimer()` / `ttSetSpeed(speed)` functions
- Update speed select `onchange` to call `ttSetSpeed(+this.value)`
- Add 64× option to speed select
- Update `_ttMovePlayhead` to also update `ttPlayheadTime` label
- Update `ttSyncBtn` to start/stop timer based on play state + speed
- Update `ttNextChunk` and `ttSeekToMs` to stop timer (restarted by onplay → ttSyncBtn)
- Rebuild Docker image after changes
8. Current Work:
Working on fixing two issues with the Timetable tab:
1. **Playback speed**: `video.playbackRate` doesn't work well for low-FPS screenpipe videos. The plan is to implement a `setInterval`-based frame-jumper for speed > 2× that advances `video.currentTime` by `speed * intervalMs / 1000` every 100–200ms, bypassing browser playbackRate limits.
2. **Timeline position indicator**: Playhead not prominent enough.
**CSS already changed** in `/volume1/screenpipe/app/static/index.html`:
```css
.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
```
Session was interrupted by summary request after this CSS edit, mid-implementation.
9. Optional Next Step:
Continuing the fix for: "still playing speed is not showing properly still just in one second. also I want to see in timeline where I am"
The next steps in order are:
1. Add `<div class="tt-playhead-time" id="ttPlayheadTime"></div>` to the HTML inside `#ttInner` (after the existing `.tt-playhead` div)
2. Update `_ttMovePlayhead` to set `ttPlayheadTime` position and text content with `fmtTime()`
3. Add timer-based speed system to JS (new functions `_ttStartSpeedTimer`, `_ttStopSpeedTimer`, `ttSetSpeed`)
4. Update speed select's `onchange` to `ttSetSpeed(+this.value)` and add 64× option
5. Update `ttSyncBtn`, `ttNextChunk`, `ttSeekToMs` to integrate with timer
6. Rebuild: `sudo docker compose build screenpipe-app && sudo docker compose up -d screenpipe-app`
If you need specific details from before compaction (like exact code snippets, error messages, or content you generated), read the full transcript at: /home/Adm1n/.claude/projects/-volume1-screenpipe/7777f46e-1e40-4051-b4bb-f43aedd59556.jsonl
Continue the conversation from where it left off without asking the user any further questions. Resume directly — do not acknowledge the summary, do not recap what was happening, do not preface with "I'll continue" or similar. Pick up the last task as if the break never happened.
Thinking
Thinking
Bash
Find timetable speed and playhead related code
IN
grep -n "tt-playhead\|ttPlayhead\|ttInner\|_ttMovePlayhead\|ttSetSpeed\|_ttStartSpeed\|_ttStopSpeed\|ttSyncBtn\|playbackRate\|speed.*select\|spdSel\|ttSpeed" /volume1/screenpipe/app/static/index.html | head -80
Copy code to clipboard
OUT
207:.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
208:.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
580: <div class="tt-inner" id="ttInner" onclick="ttTimelineClick(event)">
583: <div class="tt-playhead" id="ttPlayhead"></div>
598: onpause="ttSyncBtn()"
599: onplay="ttSyncBtn()">
607: <select id="ttSpeed" onchange="document.getElementById('ttVideo').playbackRate=+this.value">
1624: document.getElementById('ttInner').style.width = '100%';
1632: document.getElementById('ttPlayhead').style.display = 'none';
1694: document.getElementById('ttInner').style.width = (_ttZoom * 100) + '%';
1700: const inner = document.getElementById('ttInner');
1710: const inner = document.getElementById('ttInner');
1718: const inner = document.getElementById('ttInner');
1750: _ttMovePlayhead(new Date(best.ts).getTime());
1772: _ttMovePlayhead(new Date(best.ts).getTime());
1791:function _ttMovePlayhead(tsMs) {
1792: const ph = document.getElementById('ttPlayhead');
1802: const ph = document.getElementById('ttPlayhead');
1805: const inner = document.getElementById('ttInner');
1819:function ttSyncBtn() {
Read
index.html
index.html
(lines 576-625)
Read
index.html
index.html
(lines 1786-1865)
Thinking
Thinking
Bash
Find chunk loading and playhead update functions
IN
grep -n "ttUpdatePlayhead\|ttNextChunk\|ttLoadChunk\|_ttCurChunk\|_ttChunks\|_ttFrames" /volume1/screenpipe/app/static/index.html | head -50
Copy code to clipboard
OUT
596: onended="ttNextChunk()"
597: ontimeupdate="ttUpdatePlayhead()"
1569:let _ttFrames = [];
1570:let _ttChunks = [];
1571:let _ttCurChunk = -1;
1607: _ttFrames = data.frames;
1609: for (const f of _ttFrames) {
1613: _ttChunks = [...chunkMap.values()];
1614: _ttCurChunk = -1;
1616: _ttDayStart = new Date(_ttFrames[0].ts).getTime();
1617: _ttDayEnd = new Date(_ttFrames[_ttFrames.length - 1].ts).getTime();
1620: for (const f of _ttFrames) _ttColor(f.app);
1640: for (const f of _ttFrames) {
1736: if (!_ttFrames.length) return;
1737: let best = _ttFrames[0], minDiff = Infinity;
1738: for (const f of _ttFrames) {
1742: const ci = _ttChunks.findIndex(ch => ch[0].video_url === best.video_url);
1744: _ttCurChunk = ci;
1753:function ttNextChunk() {
1754: if (_ttCurChunk + 1 >= _ttChunks.length) return;
1755: _ttCurChunk++;
1757: video.src = _ttChunks[_ttCurChunk][0].video_url;
1762:function ttUpdatePlayhead() {
1764: if (_ttCurChunk < 0 || _ttCurChunk >= _ttChunks.length) return;
1765: const chunk = _ttChunks[_ttCurChunk];
Read
index.html
index.html
(lines 1754-1793)
Thinking
Thinking
Read
index.html
index.html
(lines 1701-1760)
Thinking
Thinking
Read
index.html
index.html
(lines 1561-1640)
Thinking
Thinking
Bash
Find time formatting functions
IN
grep -n "function fmtTime\|function fmtDate\|function _fmt\|localTimeToUtc\|fmtDateTime" /volume1/screenpipe/app/static/index.html | head -20
Copy code to clipboard
OUT
1088: if (tf) url += `&time_from=${localTimeToUtc(tf)}`;
1089: if (tt) url += `&time_to=${localTimeToUtc(tt)}`;
1153: let meta = `<span style="color:var(--muted)">${fmtDateTime(r.timestamp)}</span>${srcBadge}`;
1166: meta = `<span style="color:var(--muted)">${fmtDateTime(r.timestamp)}</span>${srcBadge} <span class="badge ${r.is_input_device?'mic':'sys'}">${r.is_input_device?'MIC':'SYS'}</span>`;
1307:function fmtTime(ts) {
1312:function fmtDateTime(ts) {
1320:function localTimeToUtc(timeStr) {
1778:function _fmtSec(s) {
1788: `${fmtDateTime(f.ts)} · ${f.app}${f.win ? ' / ' + f.win : ''}`;
Read
index.html
index.html
(lines 1308-1327)
Thinking
Thinking
Now I have everything I need. I'll implement the frame-by-frame timer system that replaces
video.playbackRate
with wall-clock accurate playback, plus add the playhead time label. Let me make the changes:
Edit
index.html
index.html
Added 1 line...
|
Code
|
screenpipe_sync.sh — screenpipe [SSH: nas]
|
NULL
|
|
Explorer (⇧⌘E) - 1 unsaved file
1
Search (⇧⌘F)
Explorer (⇧⌘E) - 1 unsaved file
1
Search (⇧⌘F)
Source Control (⌃⇧G) - 2 pending changes
2
Run and Debug (⇧⌘D)
Remote Explorer
Extensions (⇧⌘X) - 2 require update, 1 requires restart
3
Claude Code
Containers
EXPLORER
EXPLORER
Explorer Section: screenpipe [SSH: nas]
Explorer Section: screenpipe [SSH: nas]
SCREENPIPE [SSH: NAS]
#recycle
app
data
pipes
.gitignore
app_settings.json
archive.db
db.sqlite
screenpipe_sync.sh
screenpipe.db
Outline Section
OUTLINE
OUTLINE
Timeline Section
TIMELINE
TIMELINE
app_settings.json, Editor Group 1
screenpipe_sync.sh, Editor Group 1
db.sqlite-shm, preview, Editor Group 1
#!/bin/bash
# screenpipe_sync.sh
# Syncs Screenpipe SQLite data to a NAS archive database (append-only, no deletions).
# Also copies the day's video/frame data folder to the NAS.
#
# Usage:
# ./screenpipe_sync.sh # syncs yesterday (default)
# ./screenpipe_sync.sh 2026-04-15 # syncs a specific date
# ./screenpipe_sync.sh today # syncs today so far
#
# Cron example (runs at 3am daily):
# 0 3 * * * /Users/lukas/.screenpipe/screenpipe_sync.sh >> /Users/lukas/.screenpipe/sync.log 2>&1
set -euo pipefail
# ─── CONFIG ───────────────────────────────────────────────────────────────────
DB_SRC="${SCREENPIPE_DB:-$HOME/.screenpipe/db.sqlite}"
NAS_MOUNT="${NAS_MOUNT:-/Volumes/screenpipe}"
NAS_DB="$NAS_MOUNT/archive.db"
NAS_DATA="$NAS_MOUNT/data"
LOG_FILE="$HOME/.screenpipe/sync.log"
# ──────────────────────────────────────────────────────────────────────────────
# ─── HELPERS ──────────────────────────────────────────────────────────────────
SCRIPT_START=$(date +%s)
log() {
local msg="[$(date '+%Y-%m-%d %H:%M:%S')] $*"
echo "$msg" | tee -a "$LOG_FILE"
}
step() {
local now=$(date +%s)
local elapsed=$(( now - SCRIPT_START ))
local min=$(( elapsed / 60 ))
local sec=$(( elapsed % 60 ))
printf "\n[+%02dm%02ds] ▶ %s\n" "$min" "$sec" "$*" | tee -a "$LOG_FILE"
}
run_sqlite_heredoc() {
local label="$1"
local sql="$2"
local start=$(date +%s)
printf " %-36s " "$label"
sqlite3 "$DB_SRC" <<< "$sql" &
local pid=$!
local spin=[PASSWORD] '⠙' '⠹' '⠸' '⠼' '⠴' '⠦' '⠧' '⠇' '⠏')
local i=0
while kill -0 "$pid" 2>/dev/null; do
printf "\r %-36s %s " "$label" "${spin[$i]}"
i=$(( (i + 1) % 10 ))
sleep 0.2
done
wait "$pid"
local rc=$?
if [ $rc -ne 0 ]; then
printf "\r %-36s ✗ FAILED\n" "$label" | tee -a "$LOG_FILE"
exit $rc
fi
local dur=$(( $(date +%s) - start ))
printf "\r %-36s ✓ %dm%02ds\n" "$label" "$(( dur / 60 ))" "$(( dur % 60 ))" | tee -a "$LOG_FILE"
}
check() {
local label="$1" got="$2" expected="$3"
if [ "$got" -eq "$expected" ]; then
printf " %-20s %s / %s ✓\n" "$label:" "$got" "$expected"
else
printf " %-20s %s / %s ✗ MISMATCH\n" "$label:" "$got" "$expected"
fi
}
# ──────────────────────────────────────────────────────────────────────────────
# ─── DATE ARGUMENT ────────────────────────────────────────────────────────────
if [ "${1:-}" = "today" ]; then
TARGET_DATE=$(date +%Y-%m-%d)
elif [ -n "${1:-}" ]; then
TARGET_DATE="$1"
if ! [[ "$TARGET_DATE" =~ ^[0-9]{4}-[0-9]{2}-[0-9]{2}$ ]]; then
echo "ERROR: Invalid date format. Use YYYY-MM-DD, 'today', or no argument for yesterday."
exit 1
fi
else
TARGET_DATE=$(date -v-1d +%Y-%m-%d)
fi
log "========================================"
log "Screenpipe sync starting for: $TARGET_DATE"
log "========================================"
# ─── PREFLIGHT ────────────────────────────────────────────────────────────────
step "Preflight checks"
if [ ! -f "$DB_SRC" ]; then
log "ERROR: Source DB not found at $DB_SRC"; exit 1
fi
printf " %-20s %s (%s)\n" "Source DB:" "OK" "$(du -sh "$DB_SRC" | cut -f1)"
if [ ! -d "$NAS_MOUNT" ]; then
log "ERROR: NAS not mounted at $NAS_MOUNT"; exit 1
fi
printf " %-20s %s\n" "NAS mount:" "OK $NAS_MOUNT"
# Check if DB already synced for this date
DB_ALREADY_SYNCED=false
if [ -f "$NAS_DB" ]; then
EXISTING=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';" 2>/dev/null || echo "0")
if [ "$EXISTING" -gt "0" ]; then
log "Date $TARGET_DATE already has $EXISTING frames in archive — skipping DB sync"
DB_ALREADY_SYNCED=true
else
printf " %-20s %s (%s)\n" "Archive DB:" "exists" "$(du -sh "$NAS_DB" | cut -f1)"
fi
else
printf " %-20s %s\n" "Archive DB:" "will be created"
fi
# Source data dir for this date
DATA_SRC="$HOME/.screenpipe/data/data/$TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
DATA_SIZE=$(du -sh "$DATA_SRC" | cut -f1)
DATA_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
printf " %-20s %s (%s files, %s)\n" "Data dir:" "OK" "$DATA_FILES" "$DATA_SIZE"
else
printf " %-20s %s\n" "Data dir:" "not found — skipping file copy"
fi
# ─── DB SYNC ──────────────────────────────────────────────────────────────────
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── COUNT SOURCE ROWS ────────────────────────────────────────────────────
step "Counting source rows for $TARGET_DATE"
SRC_FRAMES=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
SRC_ELEMENTS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_UI=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
SRC_OCR=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_MEETINGS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start) = '$TARGET_DATE';")
printf " %-20s %s\n" "frames:" "$SRC_FRAMES"
printf " %-20s %s\n" "elements:" "$SRC_ELEMENTS"
printf " %-20s %s\n" "ui_events:" "$SRC_UI"
printf " %-20s %s\n" "ocr_text:" "$SRC_OCR"
printf " %-20s %s\n" "meetings:" "$SRC_MEETINGS"
if [ "$SRC_FRAMES" -eq "0" ]; then
log "No frames found for $TARGET_DATE — skipping DB sync"
DB_ALREADY_SYNCED=true
fi
fi
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── INIT TABLES ──────────────────────────────────────────────────────────
step "Initialising tables, indexes, FTS"
run_sqlite_heredoc "creating tables" "
ATTACH '$NAS_DB' AS nas;
CREATE TABLE IF NOT EXISTS nas.frames AS SELECT * FROM main.frames WHERE 0;
CREATE TABLE IF NOT EXISTS nas.elements AS SELECT * FROM main.elements WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ui_events AS SELECT * FROM main.ui_events WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ocr_text AS SELECT * FROM main.ocr_text WHERE 0;
CREATE TABLE IF NOT EXISTS nas.video_chunks AS SELECT * FROM main.video_chunks WHERE 0;
CREATE TABLE IF NOT EXISTS nas.meetings AS SELECT * FROM main.meetings WHERE 0;
DETACH nas;
"
run_sqlite_heredoc "creating indexes" "
ATTACH '$NAS_DB' AS nas;
CREATE INDEX IF NOT EXISTS nas.idx_frames_timestamp ON frames(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_frames_app_name ON frames(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_window_name ON frames(window_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_video_chunk_id ON frames(video_chunk_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_id ON elements(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_src_role ON elements(frame_id, source, role) WHERE text IS NOT NULL;
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_timestamp ON ui_events(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_app_name ON ui_events(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_frame_id ON ui_events(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_ocr_text_frame_id ON ocr_text(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_meetings_start ON meetings(meeting_start);
CREATE INDEX IF NOT EXISTS nas.idx_video_chunks_device ON video_chunks(device_name);
DETACH nas;
"
run_sqlite_heredoc "creating FTS tables" "
ATTACH '$NAS_DB' AS nas;
CREATE VIRTUAL TABLE IF NOT EXISTS nas.elements_fts USING fts5(
text, role, frame_id UNINDEXED,
content='elements', content_rowid='id', tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.frames_fts USING fts5(
full_text, app_name, window_name, browser_url, id UNINDEXED,
tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.ui_events_fts USING fts5(
text_content,
app_name,
window_title,
element_name,
content='ui_events',
content_rowid='id',
tokenize='unicode61'
);
DETACH nas;
"
# ─── SYNC DATA ────────────────────────────────────────────────────────────
step "Syncing data for $TARGET_DATE"
run_sqlite_heredoc "video_chunks" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.video_chunks
SELECT * FROM main.video_chunks
WHERE id IN (
SELECT DISTINCT video_chunk_id FROM main.frames
WHERE date(timestamp) = '$TARGET_DATE' AND video_chunk_id IS NOT NULL
);
DETACH nas;
"
run_sqlite_heredoc "frames ($SRC_FRAMES rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.frames
SELECT * FROM main.frames WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ocr_text ($SRC_OCR rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ocr_text
SELECT o.* FROM main.ocr_text o
JOIN main.frames f ON o.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ui_events ($SRC_UI rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ui_events
SELECT * FROM main.ui_events WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "elements ($SRC_ELEMENTS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.elements
SELECT e.* FROM main.elements e
JOIN main.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "meetings ($SRC_MEETINGS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.meetings
SELECT * FROM main.meetings WHERE date(meeting_start) = '$TARGET_DATE';
DETACH nas;
"
# ─── FTS UPDATE ───────────────────────────────────────────────────────────
step "Updating FTS indexes"
run_sqlite_heredoc "elements_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.elements_fts(rowid, text, role)
SELECT e.id, e.text, e.role
FROM nas.elements e
JOIN nas.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE'
AND e.text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "frames_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.frames_fts(rowid, full_text, app_name, window_name, browser_url, id)
SELECT id, full_text, app_name, window_name, browser_url, id
FROM nas.frames
WHERE date(timestamp) = '$TARGET_DATE'
AND full_text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "ui_events_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.ui_events_fts(rowid, text_content, app_name, window_title, element_name)
SELECT id, text_content, app_name, window_title, element_name
FROM nas.ui_events
WHERE date(timestamp) = '$TARGET_DATE'
AND text_content IS NOT NULL;
DETACH nas;
"
# ─── VERIFY DB ────────────────────────────────────────────────────────────
step "Verifying DB"
V_FRAMES=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
V_ELEMENTS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_UI=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
V_OCR=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_MEETINGS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start)= '$TARGET_DATE';")
check "frames" "$V_FRAMES" "$SRC_FRAMES"
check "elements" "$V_ELEMENTS" "$SRC_ELEMENTS"
check "ui_events" "$V_UI" "$SRC_UI"
check "ocr_text" "$V_OCR" "$SRC_OCR"
check "meetings" "$V_MEETINGS" "$SRC_MEETINGS"
fi
# ─── COPY DATA FOLDER ─────────────────────────────────────────────────────────
# Always runs regardless of DB sync status
step "Copying data folder for $TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
mkdir -p "$NAS_DATA/$TARGET_DATE"
RSYNC_START=$(date +%s)
printf " %-36s " "rsync $TARGET_DATE/ → NAS"
rsync -a --ignore-existing \
"$DATA_SRC/" \
"$NAS_DATA/$TARGET_DATE/" \
2>>"$LOG_FILE"
RSYNC_DUR=$(( $(date +%s) - RSYNC_START ))
COPIED_FILES=$(ls "$NAS_DATA/$TARGET_DATE" | wc -l | tr -d ' ')
SRC_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
COPIED_SIZE=$(du -sh "$NAS_DATA/$TARGET_DATE" | cut -f1)
if [ "$COPIED_FILES" -eq "$SRC_FILES" ]; then
printf "\r %-36s ✓ %dm%02ds (%s files, %s)\n" \
"rsync $TARGET_DATE/ → NAS" \
"$(( RSYNC_DUR / 60 ))" "$(( RSYNC_DUR % 60 ))" \
"$COPIED_FILES" "$COPIED_SIZE" | tee -a "$LOG_FILE"
else
printf "\r %-36s ✗ %s / %s files\n" \
"rsync $TARGET_DATE/ → NAS" "$COPIED_FILES" "$SRC_FILES" | tee -a "$LOG_FILE"
fi
else
printf " %-36s %s\n" "rsync $TARGET_DATE/ → NAS" "skipped (no source dir)"
fi
# ─── SUMMARY ──────────────────────────────────────────────────────────────────
TOTAL_ELAPSED=$(( $(date +%s) - SCRIPT_START ))
DB_SIZE=$(du -sh "$NAS_DB" | cut -f1)
echo ""
log "Archive DB size: $DB_SIZE"
log "Total time: $(( TOTAL_ELAPSED / 60 ))m$(( TOTAL_ELAPSED % 60 ))s"
log "Sync complete for $TARGET_DATE"
log "========================================"
#!/bin/bash
# screenpipe_sync.sh
# Syncs Screenpipe SQLite data to a NAS archive database (append-only, no deletions).
# Also copies the day's video/frame data folder to the NAS.
#
# Usage:
# ./screenpipe_sync.sh # syncs yesterday (default)
# ./screenpipe_sync.sh 2026-04-15 # syncs a specific date
# ./screenpipe_sync.sh today # syncs today so far
#
# Cron example (runs at 3am daily):
# 0 3 * * * /Users/lukas/.screenpipe/screenpipe_sync.sh >> /Users/lukas/.screenpipe/sync.log 2>&1
set -euo pipefail
# ─── CONFIG ───────────────────────────────────────────────────────────────────
DB_SRC="${SCREENPIPE_DB:-$HOME/.screenpipe/db.sqlite}"
NAS_MOUNT="${NAS_MOUNT:-/Volumes/screenpipe}"
NAS_DB="$NAS_MOUNT/archive.db"
NAS_DATA="$NAS_MOUNT/data"
LOG_FILE="$HOME/.screenpipe/sync.log"
# ──────────────────────────────────────────────────────────────────────────────
# ─── HELPERS ──────────────────────────────────────────────────────────────────
SCRIPT_START=$(date +%s)
log() {
local msg="[$(date '+%Y-%m-%d %H:%M:%S')] $*"
echo "$msg" | tee -a "$LOG_FILE"
}
step() {
local now=$(date +%s)
local elapsed=$(( now - SCRIPT_START ))
local min=$(( elapsed / 60 ))
local sec=$(( elapsed % 60 ))
printf "\n[+%02dm%02ds] ▶ %s\n" "$min" "$sec" "$*" | tee -a "$LOG_FILE"
}
run_sqlite_heredoc() {
local label="$1"
local sql="$2"
local start=$(date +%s)
printf " %-36s " "$label"
sqlite3 "$DB_SRC" <<< "$sql" &
local pid=$!
local spin=[PASSWORD] '⠙' '⠹' '⠸' '⠼' '⠴' '⠦' '⠧' '⠇' '⠏')
local i=0
while kill -0 "$pid" 2>/dev/null; do
printf "\r %-36s %s " "$label" "${spin[$i]}"
i=$(( (i + 1) % 10 ))
sleep 0.2
done
wait "$pid"
local rc=$?
if [ $rc -ne 0 ]; then
printf "\r %-36s ✗ FAILED\n" "$label" | tee -a "$LOG_FILE"
exit $rc
fi
local dur=$(( $(date +%s) - start ))
printf "\r %-36s ✓ %dm%02ds\n" "$label" "$(( dur / 60 ))" "$(( dur % 60 ))" | tee -a "$LOG_FILE"
}
check() {
local label="$1" got="$2" expected="$3"
if [ "$got" -eq "$expected" ]; then
printf " %-20s %s / %s ✓\n" "$label:" "$got" "$expected"
else
printf " %-20s %s / %s ✗ MISMATCH\n" "$label:" "$got" "$expected"
fi
}
# ──────────────────────────────────────────────────────────────────────────────
# ─── DATE ARGUMENT ────────────────────────────────────────────────────────────
if [ "${1:-}" = "today" ]; then
TARGET_DATE=$(date +%Y-%m-%d)
elif [ -n "${1:-}" ]; then
TARGET_DATE="$1"
if ! [[ "$TARGET_DATE" =~ ^[0-9]{4}-[0-9]{2}-[0-9]{2}$ ]]; then
echo "ERROR: Invalid date format. Use YYYY-MM-DD, 'today', or no argument for yesterday."
exit 1
fi
else
TARGET_DATE=$(date -v-1d +%Y-%m-%d)
fi
log "========================================"
log "Screenpipe sync starting for: $TARGET_DATE"
log "========================================"
# ─── PREFLIGHT ────────────────────────────────────────────────────────────────
step "Preflight checks"
if [ ! -f "$DB_SRC" ]; then
log "ERROR: Source DB not found at $DB_SRC"; exit 1
fi
printf " %-20s %s (%s)\n" "Source DB:" "OK" "$(du -sh "$DB_SRC" | cut -f1)"
if [ ! -d "$NAS_MOUNT" ]; then
log "ERROR: NAS not mounted at $NAS_MOUNT"; exit 1
fi
printf " %-20s %s\n" "NAS mount:" "OK $NAS_MOUNT"
# Check if DB already synced for this date
DB_ALREADY_SYNCED=false
if [ -f "$NAS_DB" ]; then
EXISTING=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';" 2>/dev/null || echo "0")
if [ "$EXISTING" -gt "0" ]; then
log "Date $TARGET_DATE already has $EXISTING frames in archive — skipping DB sync"
DB_ALREADY_SYNCED=true
else
printf " %-20s %s (%s)\n" "Archive DB:" "exists" "$(du -sh "$NAS_DB" | cut -f1)"
fi
else
printf " %-20s %s\n" "Archive DB:" "will be created"
fi
# Source data dir for this date
DATA_SRC="$HOME/.screenpipe/data/data/$TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
DATA_SIZE=$(du -sh "$DATA_SRC" | cut -f1)
DATA_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
printf " %-20s %s (%s files, %s)\n" "Data dir:" "OK" "$DATA_FILES" "$DATA_SIZE"
else
printf " %-20s %s\n" "Data dir:" "not found — skipping file copy"
fi
# ─── DB SYNC ──────────────────────────────────────────────────────────────────
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── COUNT SOURCE ROWS ────────────────────────────────────────────────────
step "Counting source rows for $TARGET_DATE"
SRC_FRAMES=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
SRC_ELEMENTS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_UI=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
SRC_OCR=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_MEETINGS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start) = '$TARGET_DATE';")
printf " %-20s %s\n" "frames:" "$SRC_FRAMES"
printf " %-20s %s\n" "elements:" "$SRC_ELEMENTS"
printf " %-20s %s\n" "ui_events:" "$SRC_UI"
printf " %-20s %s\n" "ocr_text:" "$SRC_OCR"
printf " %-20s %s\n" "meetings:" "$SRC_MEETINGS"
if [ "$SRC_FRAMES" -eq "0" ]; then
log "No frames found for $TARGET_DATE — skipping DB sync"
DB_ALREADY_SYNCED=true
fi
fi
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── INIT TABLES ──────────────────────────────────────────────────────────
step "Initialising tables, indexes, FTS"
run_sqlite_heredoc "creating tables" "
ATTACH '$NAS_DB' AS nas;
CREATE TABLE IF NOT EXISTS nas.frames AS SELECT * FROM main.frames WHERE 0;
CREATE TABLE IF NOT EXISTS nas.elements AS SELECT * FROM main.elements WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ui_events AS SELECT * FROM main.ui_events WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ocr_text AS SELECT * FROM main.ocr_text WHERE 0;
CREATE TABLE IF NOT EXISTS nas.video_chunks AS SELECT * FROM main.video_chunks WHERE 0;
CREATE TABLE IF NOT EXISTS nas.meetings AS SELECT * FROM main.meetings WHERE 0;
DETACH nas;
"
run_sqlite_heredoc "creating indexes" "
ATTACH '$NAS_DB' AS nas;
CREATE INDEX IF NOT EXISTS nas.idx_frames_timestamp ON frames(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_frames_app_name ON frames(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_window_name ON frames(window_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_video_chunk_id ON frames(video_chunk_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_id ON elements(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_src_role ON elements(frame_id, source, role) WHERE text IS NOT NULL;
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_timestamp ON ui_events(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_app_name ON ui_events(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_frame_id ON ui_events(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_ocr_text_frame_id ON ocr_text(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_meetings_start ON meetings(meeting_start);
CREATE INDEX IF NOT EXISTS nas.idx_video_chunks_device ON video_chunks(device_name);
DETACH nas;
"
run_sqlite_heredoc "creating FTS tables" "
ATTACH '$NAS_DB' AS nas;
CREATE VIRTUAL TABLE IF NOT EXISTS nas.elements_fts USING fts5(
text, role, frame_id UNINDEXED,
content='elements', content_rowid='id', tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.frames_fts USING fts5(
full_text, app_name, window_name, browser_url, id UNINDEXED,
tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.ui_events_fts USING fts5(
text_content,
app_name,
window_title,
element_name,
content='ui_events',
content_rowid='id',
tokenize='unicode61'
);
DETACH nas;
"
# ─── SYNC DATA ────────────────────────────────────────────────────────────
step "Syncing data for $TARGET_DATE"
run_sqlite_heredoc "video_chunks" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.video_chunks
SELECT * FROM main.video_chunks
WHERE id IN (
SELECT DISTINCT video_chunk_id FROM main.frames
WHERE date(timestamp) = '$TARGET_DATE' AND video_chunk_id IS NOT NULL
);
DETACH nas;
"
run_sqlite_heredoc "frames ($SRC_FRAMES rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.frames
SELECT * FROM main.frames WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ocr_text ($SRC_OCR rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ocr_text
SELECT o.* FROM main.ocr_text o
JOIN main.frames f ON o.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ui_events ($SRC_UI rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ui_events
SELECT * FROM main.ui_events WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "elements ($SRC_ELEMENTS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.elements
SELECT e.* FROM main.elements e
JOIN main.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "meetings ($SRC_MEETINGS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.meetings
SELECT * FROM main.meetings WHERE date(meeting_start) = '$TARGET_DATE';
DETACH nas;
"
# ─── FTS UPDATE ───────────────────────────────────────────────────────────
step "Updating FTS indexes"
run_sqlite_heredoc "elements_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.elements_fts(rowid, text, role)
SELECT e.id, e.text, e.role
FROM nas.elements e
JOIN nas.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE'
AND e.text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "frames_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.frames_fts(rowid, full_text, app_name, window_name, browser_url, id)
SELECT id, full_text, app_name, window_name, browser_url, id
FROM nas.frames
WHERE date(timestamp) = '$TARGET_DATE'
AND full_text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "ui_events_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.ui_events_fts(rowid, text_content, app_name, window_title, element_name)
SELECT id, text_content, app_name, window_title, element_name
FROM nas.ui_events
WHERE date(timestamp) = '$TARGET_DATE'
AND text_content IS NOT NULL;
DETACH nas;
"
# ─── VERIFY DB ────────────────────────────────────────────────────────────
step "Verifying DB"
V_FRAMES=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
V_ELEMENTS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_UI=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
V_OCR=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_MEETINGS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start)= '$TARGET_DATE';")
check "frames" "$V_FRAMES" "$SRC_FRAMES"
check "elements" "$V_ELEMENTS" "$SRC_ELEMENTS"
check "ui_events" "$V_UI" "$SRC_UI"
check "ocr_text" "$V_OCR" "$SRC_OCR"
check "meetings" "$V_MEETINGS" "$SRC_MEETINGS"
fi
# ─── COPY DATA FOLDER ─────────────────────────────────────────────────────────
# Always runs regardless of DB sync status
step "Copying data folder for $TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
mkdir -p "$NAS_DATA/$TARGET_DATE"
RSYNC_START=$(date +%s)
printf " %-36s " "rsync $TARGET_DATE/ → NAS"
rsync -a --ignore-existing \
"$DATA_SRC/" \
"$NAS_DATA/$TARGET_DATE/" \
2>>"$LOG_FILE"
RSYNC_DUR=$(( $(date +%s) - RSYNC_START ))
COPIED_FILES=$(ls "$NAS_DATA/$TARGET_DATE" | wc -l | tr -d ' ')
SRC_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
COPIED_SIZE=$(du -sh "$NAS_DATA/$TARGET_DATE" | cut -f1)
if [ "$COPIED_FILES" -eq "$SRC_FILES" ]; then
printf "\r %-36s ✓ %dm%02ds (%s files, %s)\n" \
"rsync $TARGET_DATE/ → NAS" \
"$(( RSYNC_DUR / 60 ))" "$(( RSYNC_DUR % 60 ))" \
"$COPIED_FILES" "$COPIED_SIZE" | tee -a "$LOG_FILE"
else
printf "\r %-36s ✗ %s / %s files\n" \
"rsync $TARGET_DATE/ → NAS" "$COPIED_FILES" "$SRC_FILES" | tee -a "$LOG_FILE"
fi
else
printf " %-36s %s\n" "rsync $TARGET_DATE/ → NAS" "skipped (no source dir)"
fi
# ─── SUMMARY ──────────────────────────────────────────────────────────────────
TOTAL_ELAPSED=$(( $(date +%s) - SCRIPT_START ))
DB_SIZE=$(du -sh "$NAS_DB" | cut -f1)
echo ""
log "Archive DB size: $DB_SIZE"
log "Total time: $(( TOTAL_ELAPSED / 60 ))m$(( TOTAL_ELAPSED % 60 ))s"
log "Sync complete for $TARGET_DATE"
log "========================================"
lets add ability to jump…, Editor Group 2
Problems (⇧⌘M)
PROBLEMS
Output (⇧⌘U)
OUTPUT
Debug Console (⇧⌘Y)
DEBUG CONSOLE
Terminal (⌃`)
TERMINAL
Ports
PORTS
remote SSH: nas
SSH: nas
screenpipe (Git) - master*, Checkout Branch/Tag...
master*
screenpipe (Git) - Synchronize Changes
No Problems
0
0
No Ports Forwarded
0
Notifications
Signed out
Signed out
Shell Script
Editor Language Status: Inline Suggestions, next: $(copilot) No inline suggestion available, Inline suggestions
LF
UTF-8
Spaces: 2
Ln 352, Col 47 (13091 selected)
Screen Reader Optimized
git-commit Lukas Kovalik (23 hours ago)
Lukas Kovalik (23 hours ago)
Info: Downloading VS Code Server...
lets add ability to jump fro,m search to timeline so we can see the exact screen (also based on monitor and time
Session history
New session
Message actions
This session is being continued from a previous conversation that ran out of context. The summary below covers the earlier portion of the conversation.
Summary:
1. Primary Request and Intent:
The user has been developing a Screenpipe dashboard on a Synology NAS. Requests in chronological order:
- **Q1**: Are db.sqlite-shm and db.sqlite-wal used? → Explained WAL mode; cleaned up files.
- **Q2**: AND combination search (term1 + term2 must both appear) → Added "AND also" input field.
- **Q3**: Explain UTC time issue, show date in results, make Only/Skip apps use multiselect → Implemented all three.
- **Q4**: Point sqlite-web at archive.db, add second sqlite-web for db.sqlite → Updated docker-compose.yml, added two services on ports 8767/8768.
- **Q5**: "still the same" (container not picking up config) → Explained `up -d` vs `restart`.
- **Q6**: Add git, push to gitea.com/lakylak/screenpipe.git → Initialized repo, committed, pushed.
- **Q7**: Add Timetable tab with video playback and timeline → Full implementation with click-to-seek, auto-advancing chunks, legend.
- **Q8**: Add scroll/zoom on timeline, jump to specific time, better awareness → Implemented zoom (1×–32×), scrollable timeline, overview minimap, viewport indicator, Jump-to-time input, screenshot thumbnail.
- **Q9 (most recent)**: Make playing speed more than 8×, fix speed not working, show position on timeline → CSS fix done (wider playhead + time label style), remainder in progress.
2. Key Technical Concepts:
- **SQLite WAL mode**: db.sqlite-shm and db.sqlite-wal are WAL companion files; cleaned with `PRAGMA wal_checkpoint(TRUNCATE)`.
- **SQLite FTS5**: Space-separated terms = AND; `MATCH 'term1 term2'` requires both; OR requires explicit `OR` keyword.
- **FastAPI**: Python backend with route handlers in main.py, consumers pattern for DB queries.
- **Docker Compose**: App image is built (code baked in), not volume-mounted → requires `build` + `up -d` for code changes. `restart` doesn't re-read compose file — need `up -d`.
- **sqlite-web**: Read-only flag `-r`, pre-built image, no rebuild needed for config changes.
- **Video playback**: screenpipe records at 0.1–0.3 fps (timelapse). `seek = offset_index / fps` gives seconds into video chunk. ~245 chunks/day across 2 monitors. Videos may be encoded at standard fps making them very short at native speed.
- **Timeline rendering**: App-colored segments built by merging consecutive frames with same `app_name`. Percentage positions: `((ts - dayStart) / span * 100)%`.
- **Zoom/scroll**: Inner div width = `zoom × 100%`, outer wrapper has `overflow-x:auto`. Playhead position stays as % of inner div, which is correct at all zoom levels.
- **Timer-based speed**: For speed > 2×, browser `playbackRate` is unreliable on low-FPS video; correct approach is `setInterval` advancing `video.currentTime` by `speed * interval_ms / 1000` seconds per tick.
- **UTC vs local time**: Backend treats `time_from`/`time_to` as UTC. Frontend converts with `localTimeToUtc()` before sending.
- **Git**: `#recycle/` in .gitignore treated as comment — must escape as `\#recycle/`.
3. Files and Code Sections:
- **`/volume1/screenpipe/app/main.py`**
- Added `/api/apps` endpoint (all unique app names from frames table)
- Added `/api/timetable` endpoint (frames + video URLs + screenshot URLs per device per date)
- `/api/timetable` response structure: `{frames: [{ts, app, win, video_url, seek, screenshot_url}], devices: [...], device: "..."}`
- Key snippet:
```python
@app.get("/api/timetable"
)
async def timetable_data(d: str | None = Query(None), device: str | None = Query(None)):
...
rows = conn.execute(f"""
SELECT f.timestamp AS ts, f.app_name AS app, f.window_name AS win,
f.offset_index, f.snapshot_path,
vc.file_path AS video_path, vc.fps AS video_fps
FROM frames f
JOIN video_chunks vc ON f.video_chunk_id = vc.id
WHERE f.timestamp >= ? AND f.timestamp < ?
AND f.video_chunk_id IS NOT NULL {dev_clause}
ORDER BY f.timestamp
""", params).fetchall()
# seek = offset_index / fps, screenshot_url from _snapshot_to_url(snapshot_path)
```
- **`/volume1/screenpipe/app/docker-compose.yml`**
- Changed sqlite-web service name to `sqlite-web-archive` → port 8767 → `/data/archive.db`
- Added `sqlite-web-live` → port 8768 → `/data/db.sqlite`
- screenpipe-app: port 8766, `DB_PATH: /data/db.sqlite`, `TZ: Europe/Sofia`
- **`/volume1/screenpipe/app/static/index.html`** (single large file with all CSS/HTML/JS)
- **Search enhancements**:
- Added `id="sq2"` "AND also" input between search-row1 and search-row2
- Replaced `sOnlyApps`/`sSkipApps` text inputs with custom multiselect components (`id="msOnlyApps"`, `id="msSkipApps"`)
- Updated `buildUrl()` to combine `q + " " + q2` and convert times with `localTimeToUtc()`
- Added `fmtDateTime(ts)` showing "Apr 20 14:08" format
- Added `localTimeToUtc(timeStr)` converting local HH:MM to UTC
- Multiselect state: `const _msState = { msOnlyApps: new Set(), msSkipApps: new Set() }`
- **Timetable tab** (new page):
- Tab button added between Work Report and AI Summary
- `showPage` array updated: `['activity','search','audio','report','timetable','summary']`
- `refresh()` calls `loadTimetable()` when on timetable page
- **Timetable CSS** (key styles):
```css
.tt-scroll-wrap { overflow-x:auto; ... scrollbar-width:thin; }
.tt-inner { position:relative; height:52px; } /* width set dynamically */
.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
.tt-overview-wrap { position:relative; height:14px; ... cursor:pointer; }
.tt-overview-vp { position:absolute; background:rgba(255,255,255,0.12); border:1px solid rgba(255,255,255,0.4); }
```
- **Timetable JS state variables**:
```javascript
let _ttFrames = [], _ttChunks = [], _ttCurChunk = -1;
let _ttDayStart = 0, _ttDayEnd = 0, _ttZoom = 1;
const _ttAppColors = {};
const TT_ZOOMS = [1, 2, 4, 8, 16, 32];
const TT_TICK_MIN = {1:30, 2:15, 4:10, 8:5, 16:2, 32:1};
```
- **Key JS functions**:
- `loadTimetable()` — fetches `/api/timetable`, groups chunks, renders
- `ttZoom(dir)` — changes zoom level, updates inner width, re-renders ticks, scrolls to playhead
- `ttTimelineClick(e)` — uses `inner.getBoundingClientRect()` (works correctly when scrolled)
- `ttOverviewClick(e)` — scrolls timeline + seeks video
- `ttUpdateOverviewVp()` — updates white viewport rect using `wrap.scrollLeft / inner.scrollWidth`
- `ttJumpToTime()` — parses time input, converts to ms, calls `ttSeekToMs`
- `ttSeekToMs(tsMs)` — binary-searches nearest frame, loads chunk, seeks video
- `_ttScrollToPlayhead(force)` — scrolls wrap so playhead stays at ~35% from left when Follow is checked
- `_ttMovePlayhead(tsMs)` — sets `ph.style.left` as percentage, calls `_ttScrollToPlayhead`
- `_ttShowNow(f)` — updates screenshot thumbnail + now-text div
- `ttSyncBtn()` — syncs play/pause button text
- Keyboard: Space=play/pause, ←/→=±10s, Ctrl+←/→=±30s, +/-=zoom
- **`/volume1/screenpipe/.gitignore`**
```
*.db, *.sqlite, *.sqlite-shm, *.sqlite-wal
data/, \#recycle/,
@eaDir/
app_settings.json, app/.env
__pycache__/, *.pyc, .DS_Store
```
4. Errors and Fixes:
- **`#recycle/` in .gitignore treated as comment**: `#` starts a comment in .gitignore. Fixed by escaping: `\#recycle/`. Also had to `git rm --cached "#recycle/desktop.ini"` to unstage it.
- **`sudo docker compose restart sqlite-web` didn't apply config change**: `restart` reuses existing container config, doesn't re-read compose file. Fix: `sudo docker compose up -d` recreates containers with new config.
- **Orphan container holding port 8767**: Old `screenpipe-sqlweb` container still running after rename. Fix: `sudo docker compose up -d --remove-orphans`.
- **`sudo docker` not accessible from Claude Code**: Interactive sudo required. User runs commands in NAS terminal.
- **`git config --global --add safe.directory /volume1/screenpipe`**: Required due to dubious ownership of NAS volume directory.
5. Problem Solving:
- **Time zone confusion in search**: Backend uses UTC for time_from/time_to; frontend displays local time. Solved by `localTimeToUtc()` converting before API call, and showing date+time in results.
- **sqlite-web not showing April 17–21 data**: Container was pointing at db.sqlite (old config), not archive.db. Fixed by updating docker-compose.yml + `up -d`.
- **Video playback speed**: Browser `video.playbackRate` is capped (~16× max in Chrome) and performs poorly for very low FPS (0.1–0.3 fps) screenpipe videos. Each chunk may only be 1–2 seconds of wall-clock encoded video, playing through instantly. Solution: implement timer-based frame-jumping for speed > 2× (in progress).
- **Playhead visibility**: White 2px line was hard to see. Fixed CSS to 3px width + stronger box-shadow + added `.tt-playhead-time` overlay label style.
6. All User Messages:
- "do we use db.sqlite-shm and db.sqlite-wal for anything right now? [image of search UI] Lets add ability to search with combination... Explain how does the search work time seems to work in utc maybe display date next to the time in results. Make only apps and skip apps multiselect with all available apps."
- "doi I need to restart?"
- "there is sudo docker compose"
- "no still the same"
- "add git. I have created gitea repo https://gitea.com/lakylak/screenpipe.git"
- "just do it" (re: git push)
- "how to make sure data in :8767 are accurate and point to archive.db"
- "I can't see older than 16th there should 21st of april. In http://[IP_ADDRESS]:8766/ I can see all but not on sqlite viewer http://[IP_ADDRESS]:8767"
- "ok lets add additional port to read db.sqlite, I would like to see the structure of official db that reads db.sqlite db and one for archive"
- [docker compose output showing orphan container error]
- "looks great. I want to have ability to scroll left and right and also to jump to specific time. I want to be more aware to what I see. Still it should be fast"
- "lets make the playing speed more than 8x"
- "still playing speed is not showing properly still just in one second. also I want to see in timeline where I am"
7. Pending Tasks:
- **Playback speed fix** (in progress, partially done):
- Add `<div class="tt-playhead-time" id="ttPlayheadTime"></div>` inside `ttInner` HTML
- Implement `_ttStartSpeedTimer(speed)` / `_ttStopSpeedTimer()` / `ttSetSpeed(speed)` functions
- Update speed select `onchange` to call `ttSetSpeed(+this.value)`
- Add 64× option to speed select
- Update `_ttMovePlayhead` to also update `ttPlayheadTime` label
- Update `ttSyncBtn` to start/stop timer based on play state + speed
- Update `ttNextChunk` and `ttSeekToMs` to stop timer (restarted by onplay → ttSyncBtn)
- Rebuild Docker image after changes
8. Current Work:
Working on fixing two issues with the Timetable tab:
1. **Playback speed**: `video.playbackRate` doesn't work well for low-FPS screenpipe videos. The plan is to implement a `setInterval`-based frame-jumper for speed > 2× that advances `video.currentTime` by `speed * intervalMs / 1000` every 100–200ms, bypassing browser playbackRate limits.
2. **Timeline position indicator**: Playhead not prominent enough.
**CSS already changed** in `/volume1/screenpipe/app/static/index.html`:
```css
.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
```
Session was interrupted by summary request after this CSS edit, mid-implementation.
9. Optional Next Step:
Continuing the fix for: "still playing speed is not showing properly still just in one second. also I want to see in timeline where I am"
The next steps in order are:
1. Add `<div class="tt-playhead-time" id="ttPlayheadTime"></div>` to the HTML inside `#ttInner` (after the existing `.tt-playhead` div)
2. Update `_ttMovePlayhead` to set `ttPlayheadTime` position and text content with `fmtTime()`
3. Add timer-based speed system to JS (new functions `_ttStartSpeedTimer`, `_ttStopSpeedTimer`, `ttSetSpeed`)
4. Update speed select's `onchange` to `ttSetSpeed(+this.value)` and add 64× option
5. Update `ttSyncBtn`, `ttNextChunk`, `ttSeekToMs` to integrate with timer
6. Rebuild: `sudo docker compose build screenpipe-app && sudo docker compose up -d screenpipe-app`
If you need specific details from before compaction (like exact code snippets, error messages, or content you generated), read the full transcript at: /home/Adm1n/.claude/projects/-volume1-screenpipe/7777f46e-1e40-4051-b4bb-f43aedd59556.jsonl
Continue the conversation from where it left off without asking the user any further questions. Resume directly — do not acknowledge the summary, do not recap what was happening, do not preface with "I'll continue" or similar. Pick up the last task as if the break never happened.
Thinking
Thinking
Bash
Find timetable speed and playhead related code
IN
grep -n "tt-playhead\|ttPlayhead\|ttInner\|_ttMovePlayhead\|ttSetSpeed\|_ttStartSpeed\|_ttStopSpeed\|ttSyncBtn\|playbackRate\|speed.*select\|spdSel\|ttSpeed" /volume1/screenpipe/app/static/index.html | head -80
Copy code to clipboard
OUT
207:.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
208:.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
580: <div class="tt-inner" id="ttInner" onclick="ttTimelineClick(event)">
583: <div class="tt-playhead" id="ttPlayhead"></div>
598: onpause="ttSyncBtn()"
599: onplay="ttSyncBtn()">
607: <select id="ttSpeed" onchange="document.getElementById('ttVideo').playbackRate=+this.value">
1624: document.getElementById('ttInner').style.width = '100%';
1632: document.getElementById('ttPlayhead').style.display = 'none';
1694: document.getElementById('ttInner').style.width = (_ttZoom * 100) + '%';
1700: const inner = document.getElementById('ttInner');
1710: const inner = document.getElementById('ttInner');
1718: const inner = document.getElementById('ttInner');
1750: _ttMovePlayhead(new Date(best.ts).getTime());
1772: _ttMovePlayhead(new Date(best.ts).getTime());
1791:function _ttMovePlayhead(tsMs) {
1792: const ph = document.getElementById('ttPlayhead');
1802: const ph = document.getElementById('ttPlayhead');
1805: const inner = document.getElementById('ttInner');
1819:function ttSyncBtn() {
Read
index.html
index.html
(lines 576-625)
Read
index.html
index.html
(lines 1786-1865)
Thinking
Thinking
Bash
Find chunk loading and playhead update functions
IN
grep -n "ttUpdatePlayhead\|ttNextChunk\|ttLoadChunk\|_ttCurChunk\|_ttChunks\|_ttFrames" /volume1/screenpipe/app/static/index.html | head -50
Copy code to clipboard
OUT
596: onended="ttNextChunk()"
597: ontimeupdate="ttUpdatePlayhead()"
1569:let _ttFrames = [];
1570:let _ttChunks = [];
1571:let _ttCurChunk = -1;
1607: _ttFrames = data.frames;
1609: for (const f of _ttFrames) {
1613: _ttChunks = [...chunkMap.values()];
1614: _ttCurChunk = -1;
1616: _ttDayStart = new Date(_ttFrames[0].ts).getTime();
1617: _ttDayEnd = new Date(_ttFrames[_ttFrames.length - 1].ts).getTime();
1620: for (const f of _ttFrames) _ttColor(f.app);
1640: for (const f of _ttFrames) {
1736: if (!_ttFrames.length) return;
1737: let best = _ttFrames[0], minDiff = Infinity;
1738: for (const f of _ttFrames) {
1742: const ci = _ttChunks.findIndex(ch => ch[0].video_url === best.video_url);
1744: _ttCurChunk = ci;
1753:function ttNextChunk() {
1754: if (_ttCurChunk + 1 >= _ttChunks.length) return;
1755: _ttCurChunk++;
1757: video.src = _ttChunks[_ttCurChunk][0].video_url;
1762:function ttUpdatePlayhead() {
1764: if (_ttCurChunk < 0 || _ttCurChunk >= _ttChunks.length) return;
1765: const chunk = _ttChunks[_ttCurChunk];
Read
index.html
index.html
(lines 1754-1793)
Thinking
Thinking
Read
index.html
index.html
(lines 1701-1760)
Thinking
Thinking
Read
index.html
index.html
(lines 1561-1640)
Thinking
Thinking
Bash
Find time formatting functions
IN
grep -n "function fmtTime\|function fmtDate\|function _fmt\|localTimeToUtc\|fmtDateTime" /volume1/screenpipe/app/static/index.html | head -20
Copy code to clipboard
OUT
1088: if (tf) url += `&time_from=${localTimeToUtc(tf)}`;
1089: if (tt) url += `&time_to=${localTimeToUtc(tt)}`;
1153: let meta = `<span style="color:var(--muted)">${fmtDateTime(r.timestamp)}</span>${srcBadge}`;
1166: meta = `<span style="color:var(--muted)">${fmtDateTime(r.timestamp)}</span>${srcBadge} <span class="badge ${r.is_input_device?'mic':'sys'}">${r.is_input_device?'MIC':'SYS'}</span>`;
1307:function fmtTime(ts) {
1312:function fmtDateTime(ts) {
1320:function localTimeToUtc(timeStr) {
1778:function _fmtSec(s) {
1788: `${fmtDateTime(f.ts)} · ${f.app}${f.win ? ' / ' + f.win : ''}`;
Read
index.html
index.html
(lines 1308-1327)
Thinking
Thinking
Now I have everything I need. I'll implement the frame-by-frame timer system that replaces
video.playbackRate
with wall-clock accurate playback, plus add the playhead time label. Let me make the changes:
Edit
index.html
index.html
Added 1 line...
|
Code
|
screenpipe_sync.sh — screenpipe [SSH: nas]
|
NULL
|
|
Explorer (⇧⌘E) - 1 unsaved file
1
Search (⇧⌘F)
Explorer (⇧⌘E) - 1 unsaved file
1
Search (⇧⌘F)
Source Control (⌃⇧G) - 2 pending changes
2
Run and Debug (⇧⌘D)
Remote Explorer
Extensions (⇧⌘X) - 2 require update, 1 requires restart
3
Claude Code
Containers
EXPLORER
EXPLORER
Explorer Section: screenpipe [SSH: nas]
Explorer Section: screenpipe [SSH: nas]
SCREENPIPE [SSH: NAS]
#recycle
app
data
pipes
.gitignore
app_settings.json
archive.db
db.sqlite
screenpipe_sync.sh
screenpipe.db
Outline Section
OUTLINE
OUTLINE
Timeline Section
TIMELINE
TIMELINE
app_settings.json, Editor Group 1
screenpipe_sync.sh, Editor Group 1
db.sqlite-shm, preview, Editor Group 1
#!/bin/bash
# screenpipe_sync.sh
# Syncs Screenpipe SQLite data to a NAS archive database (append-only, no deletions).
# Also copies the day's video/frame data folder to the NAS.
#
# Usage:
# ./screenpipe_sync.sh # syncs yesterday (default)
# ./screenpipe_sync.sh 2026-04-15 # syncs a specific date
# ./screenpipe_sync.sh today # syncs today so far
#
# Cron example (runs at 3am daily):
# 0 3 * * * /Users/lukas/.screenpipe/screenpipe_sync.sh >> /Users/lukas/.screenpipe/sync.log 2>&1
set -euo pipefail
# ─── CONFIG ───────────────────────────────────────────────────────────────────
DB_SRC="${SCREENPIPE_DB:-$HOME/.screenpipe/db.sqlite}"
NAS_MOUNT="${NAS_MOUNT:-/Volumes/screenpipe}"
NAS_DB="$NAS_MOUNT/archive.db"
NAS_DATA="$NAS_MOUNT/data"
LOG_FILE="$HOME/.screenpipe/sync.log"
# ──────────────────────────────────────────────────────────────────────────────
# ─── HELPERS ──────────────────────────────────────────────────────────────────
SCRIPT_START=$(date +%s)
log() {
local msg="[$(date '+%Y-%m-%d %H:%M:%S')] $*"
echo "$msg" | tee -a "$LOG_FILE"
}
step() {
local now=$(date +%s)
local elapsed=$(( now - SCRIPT_START ))
local min=$(( elapsed / 60 ))
local sec=$(( elapsed % 60 ))
printf "\n[+%02dm%02ds] ▶ %s\n" "$min" "$sec" "$*" | tee -a "$LOG_FILE"
}
run_sqlite_heredoc() {
local label="$1"
local sql="$2"
local start=$(date +%s)
printf " %-36s " "$label"
sqlite3 "$DB_SRC" <<< "$sql" &
local pid=$!
local spin=[PASSWORD] '⠙' '⠹' '⠸' '⠼' '⠴' '⠦' '⠧' '⠇' '⠏')
local i=0
while kill -0 "$pid" 2>/dev/null; do
printf "\r %-36s %s " "$label" "${spin[$i]}"
i=$(( (i + 1) % 10 ))
sleep 0.2
done
wait "$pid"
local rc=$?
if [ $rc -ne 0 ]; then
printf "\r %-36s ✗ FAILED\n" "$label" | tee -a "$LOG_FILE"
exit $rc
fi
local dur=$(( $(date +%s) - start ))
printf "\r %-36s ✓ %dm%02ds\n" "$label" "$(( dur / 60 ))" "$(( dur % 60 ))" | tee -a "$LOG_FILE"
}
check() {
local label="$1" got="$2" expected="$3"
if [ "$got" -eq "$expected" ]; then
printf " %-20s %s / %s ✓\n" "$label:" "$got" "$expected"
else
printf " %-20s %s / %s ✗ MISMATCH\n" "$label:" "$got" "$expected"
fi
}
# ──────────────────────────────────────────────────────────────────────────────
# ─── DATE ARGUMENT ────────────────────────────────────────────────────────────
if [ "${1:-}" = "today" ]; then
TARGET_DATE=$(date +%Y-%m-%d)
elif [ -n "${1:-}" ]; then
TARGET_DATE="$1"
if ! [[ "$TARGET_DATE" =~ ^[0-9]{4}-[0-9]{2}-[0-9]{2}$ ]]; then
echo "ERROR: Invalid date format. Use YYYY-MM-DD, 'today', or no argument for yesterday."
exit 1
fi
else
TARGET_DATE=$(date -v-1d +%Y-%m-%d)
fi
log "========================================"
log "Screenpipe sync starting for: $TARGET_DATE"
log "========================================"
# ─── PREFLIGHT ────────────────────────────────────────────────────────────────
step "Preflight checks"
if [ ! -f "$DB_SRC" ]; then
log "ERROR: Source DB not found at $DB_SRC"; exit 1
fi
printf " %-20s %s (%s)\n" "Source DB:" "OK" "$(du -sh "$DB_SRC" | cut -f1)"
if [ ! -d "$NAS_MOUNT" ]; then
log "ERROR: NAS not mounted at $NAS_MOUNT"; exit 1
fi
printf " %-20s %s\n" "NAS mount:" "OK $NAS_MOUNT"
# Check if DB already synced for this date
DB_ALREADY_SYNCED=false
if [ -f "$NAS_DB" ]; then
EXISTING=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';" 2>/dev/null || echo "0")
if [ "$EXISTING" -gt "0" ]; then
log "Date $TARGET_DATE already has $EXISTING frames in archive — skipping DB sync"
DB_ALREADY_SYNCED=true
else
printf " %-20s %s (%s)\n" "Archive DB:" "exists" "$(du -sh "$NAS_DB" | cut -f1)"
fi
else
printf " %-20s %s\n" "Archive DB:" "will be created"
fi
# Source data dir for this date
DATA_SRC="$HOME/.screenpipe/data/data/$TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
DATA_SIZE=$(du -sh "$DATA_SRC" | cut -f1)
DATA_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
printf " %-20s %s (%s files, %s)\n" "Data dir:" "OK" "$DATA_FILES" "$DATA_SIZE"
else
printf " %-20s %s\n" "Data dir:" "not found — skipping file copy"
fi
# ─── DB SYNC ──────────────────────────────────────────────────────────────────
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── COUNT SOURCE ROWS ────────────────────────────────────────────────────
step "Counting source rows for $TARGET_DATE"
SRC_FRAMES=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
SRC_ELEMENTS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_UI=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
SRC_OCR=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_MEETINGS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start) = '$TARGET_DATE';")
printf " %-20s %s\n" "frames:" "$SRC_FRAMES"
printf " %-20s %s\n" "elements:" "$SRC_ELEMENTS"
printf " %-20s %s\n" "ui_events:" "$SRC_UI"
printf " %-20s %s\n" "ocr_text:" "$SRC_OCR"
printf " %-20s %s\n" "meetings:" "$SRC_MEETINGS"
if [ "$SRC_FRAMES" -eq "0" ]; then
log "No frames found for $TARGET_DATE — skipping DB sync"
DB_ALREADY_SYNCED=true
fi
fi
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── INIT TABLES ──────────────────────────────────────────────────────────
step "Initialising tables, indexes, FTS"
run_sqlite_heredoc "creating tables" "
ATTACH '$NAS_DB' AS nas;
CREATE TABLE IF NOT EXISTS nas.frames AS SELECT * FROM main.frames WHERE 0;
CREATE TABLE IF NOT EXISTS nas.elements AS SELECT * FROM main.elements WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ui_events AS SELECT * FROM main.ui_events WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ocr_text AS SELECT * FROM main.ocr_text WHERE 0;
CREATE TABLE IF NOT EXISTS nas.video_chunks AS SELECT * FROM main.video_chunks WHERE 0;
CREATE TABLE IF NOT EXISTS nas.meetings AS SELECT * FROM main.meetings WHERE 0;
DETACH nas;
"
run_sqlite_heredoc "creating indexes" "
ATTACH '$NAS_DB' AS nas;
CREATE INDEX IF NOT EXISTS nas.idx_frames_timestamp ON frames(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_frames_app_name ON frames(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_window_name ON frames(window_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_video_chunk_id ON frames(video_chunk_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_id ON elements(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_src_role ON elements(frame_id, source, role) WHERE text IS NOT NULL;
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_timestamp ON ui_events(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_app_name ON ui_events(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_frame_id ON ui_events(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_ocr_text_frame_id ON ocr_text(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_meetings_start ON meetings(meeting_start);
CREATE INDEX IF NOT EXISTS nas.idx_video_chunks_device ON video_chunks(device_name);
DETACH nas;
"
run_sqlite_heredoc "creating FTS tables" "
ATTACH '$NAS_DB' AS nas;
CREATE VIRTUAL TABLE IF NOT EXISTS nas.elements_fts USING fts5(
text, role, frame_id UNINDEXED,
content='elements', content_rowid='id', tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.frames_fts USING fts5(
full_text, app_name, window_name, browser_url, id UNINDEXED,
tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.ui_events_fts USING fts5(
text_content,
app_name,
window_title,
element_name,
content='ui_events',
content_rowid='id',
tokenize='unicode61'
);
DETACH nas;
"
# ─── SYNC DATA ────────────────────────────────────────────────────────────
step "Syncing data for $TARGET_DATE"
run_sqlite_heredoc "video_chunks" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.video_chunks
SELECT * FROM main.video_chunks
WHERE id IN (
SELECT DISTINCT video_chunk_id FROM main.frames
WHERE date(timestamp) = '$TARGET_DATE' AND video_chunk_id IS NOT NULL
);
DETACH nas;
"
run_sqlite_heredoc "frames ($SRC_FRAMES rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.frames
SELECT * FROM main.frames WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ocr_text ($SRC_OCR rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ocr_text
SELECT o.* FROM main.ocr_text o
JOIN main.frames f ON o.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ui_events ($SRC_UI rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ui_events
SELECT * FROM main.ui_events WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "elements ($SRC_ELEMENTS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.elements
SELECT e.* FROM main.elements e
JOIN main.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "meetings ($SRC_MEETINGS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.meetings
SELECT * FROM main.meetings WHERE date(meeting_start) = '$TARGET_DATE';
DETACH nas;
"
# ─── FTS UPDATE ───────────────────────────────────────────────────────────
step "Updating FTS indexes"
run_sqlite_heredoc "elements_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.elements_fts(rowid, text, role)
SELECT e.id, e.text, e.role
FROM nas.elements e
JOIN nas.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE'
AND e.text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "frames_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.frames_fts(rowid, full_text, app_name, window_name, browser_url, id)
SELECT id, full_text, app_name, window_name, browser_url, id
FROM nas.frames
WHERE date(timestamp) = '$TARGET_DATE'
AND full_text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "ui_events_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.ui_events_fts(rowid, text_content, app_name, window_title, element_name)
SELECT id, text_content, app_name, window_title, element_name
FROM nas.ui_events
WHERE date(timestamp) = '$TARGET_DATE'
AND text_content IS NOT NULL;
DETACH nas;
"
# ─── VERIFY DB ────────────────────────────────────────────────────────────
step "Verifying DB"
V_FRAMES=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
V_ELEMENTS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_UI=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
V_OCR=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_MEETINGS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start)= '$TARGET_DATE';")
check "frames" "$V_FRAMES" "$SRC_FRAMES"
check "elements" "$V_ELEMENTS" "$SRC_ELEMENTS"
check "ui_events" "$V_UI" "$SRC_UI"
check "ocr_text" "$V_OCR" "$SRC_OCR"
check "meetings" "$V_MEETINGS" "$SRC_MEETINGS"
fi
# ─── COPY DATA FOLDER ─────────────────────────────────────────────────────────
# Always runs regardless of DB sync status
step "Copying data folder for $TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
mkdir -p "$NAS_DATA/$TARGET_DATE"
RSYNC_START=$(date +%s)
printf " %-36s " "rsync $TARGET_DATE/ → NAS"
rsync -a --ignore-existing \
"$DATA_SRC/" \
"$NAS_DATA/$TARGET_DATE/" \
2>>"$LOG_FILE"
RSYNC_DUR=$(( $(date +%s) - RSYNC_START ))
COPIED_FILES=$(ls "$NAS_DATA/$TARGET_DATE" | wc -l | tr -d ' ')
SRC_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
COPIED_SIZE=$(du -sh "$NAS_DATA/$TARGET_DATE" | cut -f1)
if [ "$COPIED_FILES" -eq "$SRC_FILES" ]; then
printf "\r %-36s ✓ %dm%02ds (%s files, %s)\n" \
"rsync $TARGET_DATE/ → NAS" \
"$(( RSYNC_DUR / 60 ))" "$(( RSYNC_DUR % 60 ))" \
"$COPIED_FILES" "$COPIED_SIZE" | tee -a "$LOG_FILE"
else
printf "\r %-36s ✗ %s / %s files\n" \
"rsync $TARGET_DATE/ → NAS" "$COPIED_FILES" "$SRC_FILES" | tee -a "$LOG_FILE"
fi
else
printf " %-36s %s\n" "rsync $TARGET_DATE/ → NAS" "skipped (no source dir)"
fi
# ─── SUMMARY ──────────────────────────────────────────────────────────────────
TOTAL_ELAPSED=$(( $(date +%s) - SCRIPT_START ))
DB_SIZE=$(du -sh "$NAS_DB" | cut -f1)
echo ""
log "Archive DB size: $DB_SIZE"
log "Total time: $(( TOTAL_ELAPSED / 60 ))m$(( TOTAL_ELAPSED % 60 ))s"
log "Sync complete for $TARGET_DATE"
log "========================================"
#!/bin/bash
# screenpipe_sync.sh
# Syncs Screenpipe SQLite data to a NAS archive database (append-only, no deletions).
# Also copies the day's video/frame data folder to the NAS.
#
# Usage:
# ./screenpipe_sync.sh # syncs yesterday (default)
# ./screenpipe_sync.sh 2026-04-15 # syncs a specific date
# ./screenpipe_sync.sh today # syncs today so far
#
# Cron example (runs at 3am daily):
# 0 3 * * * /Users/lukas/.screenpipe/screenpipe_sync.sh >> /Users/lukas/.screenpipe/sync.log 2>&1
set -euo pipefail
# ─── CONFIG ───────────────────────────────────────────────────────────────────
DB_SRC="${SCREENPIPE_DB:-$HOME/.screenpipe/db.sqlite}"
NAS_MOUNT="${NAS_MOUNT:-/Volumes/screenpipe}"
NAS_DB="$NAS_MOUNT/archive.db"
NAS_DATA="$NAS_MOUNT/data"
LOG_FILE="$HOME/.screenpipe/sync.log"
# ──────────────────────────────────────────────────────────────────────────────
# ─── HELPERS ──────────────────────────────────────────────────────────────────
SCRIPT_START=$(date +%s)
log() {
local msg="[$(date '+%Y-%m-%d %H:%M:%S')] $*"
echo "$msg" | tee -a "$LOG_FILE"
}
step() {
local now=$(date +%s)
local elapsed=$(( now - SCRIPT_START ))
local min=$(( elapsed / 60 ))
local sec=$(( elapsed % 60 ))
printf "\n[+%02dm%02ds] ▶ %s\n" "$min" "$sec" "$*" | tee -a "$LOG_FILE"
}
run_sqlite_heredoc() {
local label="$1"
local sql="$2"
local start=$(date +%s)
printf " %-36s " "$label"
sqlite3 "$DB_SRC" <<< "$sql" &
local pid=$!
local spin=[PASSWORD] '⠙' '⠹' '⠸' '⠼' '⠴' '⠦' '⠧' '⠇' '⠏')
local i=0
while kill -0 "$pid" 2>/dev/null; do
printf "\r %-36s %s " "$label" "${spin[$i]}"
i=$(( (i + 1) % 10 ))
sleep 0.2
done
wait "$pid"
local rc=$?
if [ $rc -ne 0 ]; then
printf "\r %-36s ✗ FAILED\n" "$label" | tee -a "$LOG_FILE"
exit $rc
fi
local dur=$(( $(date +%s) - start ))
printf "\r %-36s ✓ %dm%02ds\n" "$label" "$(( dur / 60 ))" "$(( dur % 60 ))" | tee -a "$LOG_FILE"
}
check() {
local label="$1" got="$2" expected="$3"
if [ "$got" -eq "$expected" ]; then
printf " %-20s %s / %s ✓\n" "$label:" "$got" "$expected"
else
printf " %-20s %s / %s ✗ MISMATCH\n" "$label:" "$got" "$expected"
fi
}
# ──────────────────────────────────────────────────────────────────────────────
# ─── DATE ARGUMENT ────────────────────────────────────────────────────────────
if [ "${1:-}" = "today" ]; then
TARGET_DATE=$(date +%Y-%m-%d)
elif [ -n "${1:-}" ]; then
TARGET_DATE="$1"
if ! [[ "$TARGET_DATE" =~ ^[0-9]{4}-[0-9]{2}-[0-9]{2}$ ]]; then
echo "ERROR: Invalid date format. Use YYYY-MM-DD, 'today', or no argument for yesterday."
exit 1
fi
else
TARGET_DATE=$(date -v-1d +%Y-%m-%d)
fi
log "========================================"
log "Screenpipe sync starting for: $TARGET_DATE"
log "========================================"
# ─── PREFLIGHT ────────────────────────────────────────────────────────────────
step "Preflight checks"
if [ ! -f "$DB_SRC" ]; then
log "ERROR: Source DB not found at $DB_SRC"; exit 1
fi
printf " %-20s %s (%s)\n" "Source DB:" "OK" "$(du -sh "$DB_SRC" | cut -f1)"
if [ ! -d "$NAS_MOUNT" ]; then
log "ERROR: NAS not mounted at $NAS_MOUNT"; exit 1
fi
printf " %-20s %s\n" "NAS mount:" "OK $NAS_MOUNT"
# Check if DB already synced for this date
DB_ALREADY_SYNCED=false
if [ -f "$NAS_DB" ]; then
EXISTING=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';" 2>/dev/null || echo "0")
if [ "$EXISTING" -gt "0" ]; then
log "Date $TARGET_DATE already has $EXISTING frames in archive — skipping DB sync"
DB_ALREADY_SYNCED=true
else
printf " %-20s %s (%s)\n" "Archive DB:" "exists" "$(du -sh "$NAS_DB" | cut -f1)"
fi
else
printf " %-20s %s\n" "Archive DB:" "will be created"
fi
# Source data dir for this date
DATA_SRC="$HOME/.screenpipe/data/data/$TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
DATA_SIZE=$(du -sh "$DATA_SRC" | cut -f1)
DATA_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
printf " %-20s %s (%s files, %s)\n" "Data dir:" "OK" "$DATA_FILES" "$DATA_SIZE"
else
printf " %-20s %s\n" "Data dir:" "not found — skipping file copy"
fi
# ─── DB SYNC ──────────────────────────────────────────────────────────────────
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── COUNT SOURCE ROWS ────────────────────────────────────────────────────
step "Counting source rows for $TARGET_DATE"
SRC_FRAMES=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
SRC_ELEMENTS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_UI=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
SRC_OCR=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
SRC_MEETINGS=$(sqlite3 "$DB_SRC" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start) = '$TARGET_DATE';")
printf " %-20s %s\n" "frames:" "$SRC_FRAMES"
printf " %-20s %s\n" "elements:" "$SRC_ELEMENTS"
printf " %-20s %s\n" "ui_events:" "$SRC_UI"
printf " %-20s %s\n" "ocr_text:" "$SRC_OCR"
printf " %-20s %s\n" "meetings:" "$SRC_MEETINGS"
if [ "$SRC_FRAMES" -eq "0" ]; then
log "No frames found for $TARGET_DATE — skipping DB sync"
DB_ALREADY_SYNCED=true
fi
fi
if [ "$DB_ALREADY_SYNCED" = false ]; then
# ─── INIT TABLES ──────────────────────────────────────────────────────────
step "Initialising tables, indexes, FTS"
run_sqlite_heredoc "creating tables" "
ATTACH '$NAS_DB' AS nas;
CREATE TABLE IF NOT EXISTS nas.frames AS SELECT * FROM main.frames WHERE 0;
CREATE TABLE IF NOT EXISTS nas.elements AS SELECT * FROM main.elements WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ui_events AS SELECT * FROM main.ui_events WHERE 0;
CREATE TABLE IF NOT EXISTS nas.ocr_text AS SELECT * FROM main.ocr_text WHERE 0;
CREATE TABLE IF NOT EXISTS nas.video_chunks AS SELECT * FROM main.video_chunks WHERE 0;
CREATE TABLE IF NOT EXISTS nas.meetings AS SELECT * FROM main.meetings WHERE 0;
DETACH nas;
"
run_sqlite_heredoc "creating indexes" "
ATTACH '$NAS_DB' AS nas;
CREATE INDEX IF NOT EXISTS nas.idx_frames_timestamp ON frames(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_frames_app_name ON frames(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_window_name ON frames(window_name);
CREATE INDEX IF NOT EXISTS nas.idx_frames_video_chunk_id ON frames(video_chunk_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_id ON elements(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_elements_frame_src_role ON elements(frame_id, source, role) WHERE text IS NOT NULL;
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_timestamp ON ui_events(timestamp);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_app_name ON ui_events(app_name);
CREATE INDEX IF NOT EXISTS nas.idx_ui_events_frame_id ON ui_events(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_ocr_text_frame_id ON ocr_text(frame_id);
CREATE INDEX IF NOT EXISTS nas.idx_meetings_start ON meetings(meeting_start);
CREATE INDEX IF NOT EXISTS nas.idx_video_chunks_device ON video_chunks(device_name);
DETACH nas;
"
run_sqlite_heredoc "creating FTS tables" "
ATTACH '$NAS_DB' AS nas;
CREATE VIRTUAL TABLE IF NOT EXISTS nas.elements_fts USING fts5(
text, role, frame_id UNINDEXED,
content='elements', content_rowid='id', tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.frames_fts USING fts5(
full_text, app_name, window_name, browser_url, id UNINDEXED,
tokenize='unicode61'
);
CREATE VIRTUAL TABLE IF NOT EXISTS nas.ui_events_fts USING fts5(
text_content,
app_name,
window_title,
element_name,
content='ui_events',
content_rowid='id',
tokenize='unicode61'
);
DETACH nas;
"
# ─── SYNC DATA ────────────────────────────────────────────────────────────
step "Syncing data for $TARGET_DATE"
run_sqlite_heredoc "video_chunks" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.video_chunks
SELECT * FROM main.video_chunks
WHERE id IN (
SELECT DISTINCT video_chunk_id FROM main.frames
WHERE date(timestamp) = '$TARGET_DATE' AND video_chunk_id IS NOT NULL
);
DETACH nas;
"
run_sqlite_heredoc "frames ($SRC_FRAMES rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.frames
SELECT * FROM main.frames WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ocr_text ($SRC_OCR rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ocr_text
SELECT o.* FROM main.ocr_text o
JOIN main.frames f ON o.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "ui_events ($SRC_UI rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.ui_events
SELECT * FROM main.ui_events WHERE date(timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "elements ($SRC_ELEMENTS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.elements
SELECT e.* FROM main.elements e
JOIN main.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE';
DETACH nas;
"
run_sqlite_heredoc "meetings ($SRC_MEETINGS rows)" "
ATTACH '$NAS_DB' AS nas;
INSERT OR IGNORE INTO nas.meetings
SELECT * FROM main.meetings WHERE date(meeting_start) = '$TARGET_DATE';
DETACH nas;
"
# ─── FTS UPDATE ───────────────────────────────────────────────────────────
step "Updating FTS indexes"
run_sqlite_heredoc "elements_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.elements_fts(rowid, text, role)
SELECT e.id, e.text, e.role
FROM nas.elements e
JOIN nas.frames f ON e.frame_id = f.id
WHERE date(f.timestamp) = '$TARGET_DATE'
AND e.text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "frames_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.frames_fts(rowid, full_text, app_name, window_name, browser_url, id)
SELECT id, full_text, app_name, window_name, browser_url, id
FROM nas.frames
WHERE date(timestamp) = '$TARGET_DATE'
AND full_text IS NOT NULL;
DETACH nas;
"
run_sqlite_heredoc "ui_events_fts" "
ATTACH '$NAS_DB' AS nas;
INSERT INTO nas.ui_events_fts(rowid, text_content, app_name, window_title, element_name)
SELECT id, text_content, app_name, window_title, element_name
FROM nas.ui_events
WHERE date(timestamp) = '$TARGET_DATE'
AND text_content IS NOT NULL;
DETACH nas;
"
# ─── VERIFY DB ────────────────────────────────────────────────────────────
step "Verifying DB"
V_FRAMES=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM frames WHERE date(timestamp) = '$TARGET_DATE';")
V_ELEMENTS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM elements WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_UI=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ui_events WHERE date(timestamp) = '$TARGET_DATE';")
V_OCR=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM ocr_text WHERE frame_id IN (SELECT id FROM frames WHERE date(timestamp) = '$TARGET_DATE');")
V_MEETINGS=$(sqlite3 "$NAS_DB" "SELECT COUNT(*) FROM meetings WHERE date(meeting_start)= '$TARGET_DATE';")
check "frames" "$V_FRAMES" "$SRC_FRAMES"
check "elements" "$V_ELEMENTS" "$SRC_ELEMENTS"
check "ui_events" "$V_UI" "$SRC_UI"
check "ocr_text" "$V_OCR" "$SRC_OCR"
check "meetings" "$V_MEETINGS" "$SRC_MEETINGS"
fi
# ─── COPY DATA FOLDER ─────────────────────────────────────────────────────────
# Always runs regardless of DB sync status
step "Copying data folder for $TARGET_DATE"
if [ -d "$DATA_SRC" ]; then
mkdir -p "$NAS_DATA/$TARGET_DATE"
RSYNC_START=$(date +%s)
printf " %-36s " "rsync $TARGET_DATE/ → NAS"
rsync -a --ignore-existing \
"$DATA_SRC/" \
"$NAS_DATA/$TARGET_DATE/" \
2>>"$LOG_FILE"
RSYNC_DUR=$(( $(date +%s) - RSYNC_START ))
COPIED_FILES=$(ls "$NAS_DATA/$TARGET_DATE" | wc -l | tr -d ' ')
SRC_FILES=$(ls "$DATA_SRC" | wc -l | tr -d ' ')
COPIED_SIZE=$(du -sh "$NAS_DATA/$TARGET_DATE" | cut -f1)
if [ "$COPIED_FILES" -eq "$SRC_FILES" ]; then
printf "\r %-36s ✓ %dm%02ds (%s files, %s)\n" \
"rsync $TARGET_DATE/ → NAS" \
"$(( RSYNC_DUR / 60 ))" "$(( RSYNC_DUR % 60 ))" \
"$COPIED_FILES" "$COPIED_SIZE" | tee -a "$LOG_FILE"
else
printf "\r %-36s ✗ %s / %s files\n" \
"rsync $TARGET_DATE/ → NAS" "$COPIED_FILES" "$SRC_FILES" | tee -a "$LOG_FILE"
fi
else
printf " %-36s %s\n" "rsync $TARGET_DATE/ → NAS" "skipped (no source dir)"
fi
# ─── SUMMARY ──────────────────────────────────────────────────────────────────
TOTAL_ELAPSED=$(( $(date +%s) - SCRIPT_START ))
DB_SIZE=$(du -sh "$NAS_DB" | cut -f1)
echo ""
log "Archive DB size: $DB_SIZE"
log "Total time: $(( TOTAL_ELAPSED / 60 ))m$(( TOTAL_ELAPSED % 60 ))s"
log "Sync complete for $TARGET_DATE"
log "========================================"
lets add ability to jump…, Editor Group 2
Problems (⇧⌘M)
PROBLEMS
Output (⇧⌘U)
OUTPUT
Debug Console (⇧⌘Y)
DEBUG CONSOLE
Terminal (⌃`)
TERMINAL
Ports
PORTS
remote SSH: nas
SSH: nas
screenpipe (Git) - master*, Checkout Branch/Tag...
master*
screenpipe (Git) - Synchronize Changes
No Problems
0
0
No Ports Forwarded
0
Notifications
Signed out
Signed out
Shell Script
Editor Language Status: Inline Suggestions, next: $(copilot) No inline suggestion available, Inline suggestions
LF
UTF-8
Spaces: 2
Ln 352, Col 47 (13091 selected)
Screen Reader Optimized
git-commit Lukas Kovalik (23 hours ago)
Lukas Kovalik (23 hours ago)
Info: Downloading VS Code Server...
lets add ability to jump fro,m search to timeline so we can see the exact screen (also based on monitor and time
Session history
New session
Message actions
This session is being continued from a previous conversation that ran out of context. The summary below covers the earlier portion of the conversation.
Summary:
1. Primary Request and Intent:
The user has been developing a Screenpipe dashboard on a Synology NAS. Requests in chronological order:
- **Q1**: Are db.sqlite-shm and db.sqlite-wal used? → Explained WAL mode; cleaned up files.
- **Q2**: AND combination search (term1 + term2 must both appear) → Added "AND also" input field.
- **Q3**: Explain UTC time issue, show date in results, make Only/Skip apps use multiselect → Implemented all three.
- **Q4**: Point sqlite-web at archive.db, add second sqlite-web for db.sqlite → Updated docker-compose.yml, added two services on ports 8767/8768.
- **Q5**: "still the same" (container not picking up config) → Explained `up -d` vs `restart`.
- **Q6**: Add git, push to gitea.com/lakylak/screenpipe.git → Initialized repo, committed, pushed.
- **Q7**: Add Timetable tab with video playback and timeline → Full implementation with click-to-seek, auto-advancing chunks, legend.
- **Q8**: Add scroll/zoom on timeline, jump to specific time, better awareness → Implemented zoom (1×–32×), scrollable timeline, overview minimap, viewport indicator, Jump-to-time input, screenshot thumbnail.
- **Q9 (most recent)**: Make playing speed more than 8×, fix speed not working, show position on timeline → CSS fix done (wider playhead + time label style), remainder in progress.
2. Key Technical Concepts:
- **SQLite WAL mode**: db.sqlite-shm and db.sqlite-wal are WAL companion files; cleaned with `PRAGMA wal_checkpoint(TRUNCATE)`.
- **SQLite FTS5**: Space-separated terms = AND; `MATCH 'term1 term2'` requires both; OR requires explicit `OR` keyword.
- **FastAPI**: Python backend with route handlers in main.py, consumers pattern for DB queries.
- **Docker Compose**: App image is built (code baked in), not volume-mounted → requires `build` + `up -d` for code changes. `restart` doesn't re-read compose file — need `up -d`.
- **sqlite-web**: Read-only flag `-r`, pre-built image, no rebuild needed for config changes.
- **Video playback**: screenpipe records at 0.1–0.3 fps (timelapse). `seek = offset_index / fps` gives seconds into video chunk. ~245 chunks/day across 2 monitors. Videos may be encoded at standard fps making them very short at native speed.
- **Timeline rendering**: App-colored segments built by merging consecutive frames with same `app_name`. Percentage positions: `((ts - dayStart) / span * 100)%`.
- **Zoom/scroll**: Inner div width = `zoom × 100%`, outer wrapper has `overflow-x:auto`. Playhead position stays as % of inner div, which is correct at all zoom levels.
- **Timer-based speed**: For speed > 2×, browser `playbackRate` is unreliable on low-FPS video; correct approach is `setInterval` advancing `video.currentTime` by `speed * interval_ms / 1000` seconds per tick.
- **UTC vs local time**: Backend treats `time_from`/`time_to` as UTC. Frontend converts with `localTimeToUtc()` before sending.
- **Git**: `#recycle/` in .gitignore treated as comment — must escape as `\#recycle/`.
3. Files and Code Sections:
- **`/volume1/screenpipe/app/main.py`**
- Added `/api/apps` endpoint (all unique app names from frames table)
- Added `/api/timetable` endpoint (frames + video URLs + screenshot URLs per device per date)
- `/api/timetable` response structure: `{frames: [{ts, app, win, video_url, seek, screenshot_url}], devices: [...], device: "..."}`
- Key snippet:
```python
@app.get("/api/timetable"
)
async def timetable_data(d: str | None = Query(None), device: str | None = Query(None)):
...
rows = conn.execute(f"""
SELECT f.timestamp AS ts, f.app_name AS app, f.window_name AS win,
f.offset_index, f.snapshot_path,
vc.file_path AS video_path, vc.fps AS video_fps
FROM frames f
JOIN video_chunks vc ON f.video_chunk_id = vc.id
WHERE f.timestamp >= ? AND f.timestamp < ?
AND f.video_chunk_id IS NOT NULL {dev_clause}
ORDER BY f.timestamp
""", params).fetchall()
# seek = offset_index / fps, screenshot_url from _snapshot_to_url(snapshot_path)
```
- **`/volume1/screenpipe/app/docker-compose.yml`**
- Changed sqlite-web service name to `sqlite-web-archive` → port 8767 → `/data/archive.db`
- Added `sqlite-web-live` → port 8768 → `/data/db.sqlite`
- screenpipe-app: port 8766, `DB_PATH: /data/db.sqlite`, `TZ: Europe/Sofia`
- **`/volume1/screenpipe/app/static/index.html`** (single large file with all CSS/HTML/JS)
- **Search enhancements**:
- Added `id="sq2"` "AND also" input between search-row1 and search-row2
- Replaced `sOnlyApps`/`sSkipApps` text inputs with custom multiselect components (`id="msOnlyApps"`, `id="msSkipApps"`)
- Updated `buildUrl()` to combine `q + " " + q2` and convert times with `localTimeToUtc()`
- Added `fmtDateTime(ts)` showing "Apr 20 14:08" format
- Added `localTimeToUtc(timeStr)` converting local HH:MM to UTC
- Multiselect state: `const _msState = { msOnlyApps: new Set(), msSkipApps: new Set() }`
- **Timetable tab** (new page):
- Tab button added between Work Report and AI Summary
- `showPage` array updated: `['activity','search','audio','report','timetable','summary']`
- `refresh()` calls `loadTimetable()` when on timetable page
- **Timetable CSS** (key styles):
```css
.tt-scroll-wrap { overflow-x:auto; ... scrollbar-width:thin; }
.tt-inner { position:relative; height:52px; } /* width set dynamically */
.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
.tt-overview-wrap { position:relative; height:14px; ... cursor:pointer; }
.tt-overview-vp { position:absolute; background:rgba(255,255,255,0.12); border:1px solid rgba(255,255,255,0.4); }
```
- **Timetable JS state variables**:
```javascript
let _ttFrames = [], _ttChunks = [], _ttCurChunk = -1;
let _ttDayStart = 0, _ttDayEnd = 0, _ttZoom = 1;
const _ttAppColors = {};
const TT_ZOOMS = [1, 2, 4, 8, 16, 32];
const TT_TICK_MIN = {1:30, 2:15, 4:10, 8:5, 16:2, 32:1};
```
- **Key JS functions**:
- `loadTimetable()` — fetches `/api/timetable`, groups chunks, renders
- `ttZoom(dir)` — changes zoom level, updates inner width, re-renders ticks, scrolls to playhead
- `ttTimelineClick(e)` — uses `inner.getBoundingClientRect()` (works correctly when scrolled)
- `ttOverviewClick(e)` — scrolls timeline + seeks video
- `ttUpdateOverviewVp()` — updates white viewport rect using `wrap.scrollLeft / inner.scrollWidth`
- `ttJumpToTime()` — parses time input, converts to ms, calls `ttSeekToMs`
- `ttSeekToMs(tsMs)` — binary-searches nearest frame, loads chunk, seeks video
- `_ttScrollToPlayhead(force)` — scrolls wrap so playhead stays at ~35% from left when Follow is checked
- `_ttMovePlayhead(tsMs)` — sets `ph.style.left` as percentage, calls `_ttScrollToPlayhead`
- `_ttShowNow(f)` — updates screenshot thumbnail + now-text div
- `ttSyncBtn()` — syncs play/pause button text
- Keyboard: Space=play/pause, ←/→=±10s, Ctrl+←/→=±30s, +/-=zoom
- **`/volume1/screenpipe/.gitignore`**
```
*.db, *.sqlite, *.sqlite-shm, *.sqlite-wal
data/, \#recycle/,
@eaDir/
app_settings.json, app/.env
__pycache__/, *.pyc, .DS_Store
```
4. Errors and Fixes:
- **`#recycle/` in .gitignore treated as comment**: `#` starts a comment in .gitignore. Fixed by escaping: `\#recycle/`. Also had to `git rm --cached "#recycle/desktop.ini"` to unstage it.
- **`sudo docker compose restart sqlite-web` didn't apply config change**: `restart` reuses existing container config, doesn't re-read compose file. Fix: `sudo docker compose up -d` recreates containers with new config.
- **Orphan container holding port 8767**: Old `screenpipe-sqlweb` container still running after rename. Fix: `sudo docker compose up -d --remove-orphans`.
- **`sudo docker` not accessible from Claude Code**: Interactive sudo required. User runs commands in NAS terminal.
- **`git config --global --add safe.directory /volume1/screenpipe`**: Required due to dubious ownership of NAS volume directory.
5. Problem Solving:
- **Time zone confusion in search**: Backend uses UTC for time_from/time_to; frontend displays local time. Solved by `localTimeToUtc()` converting before API call, and showing date+time in results.
- **sqlite-web not showing April 17–21 data**: Container was pointing at db.sqlite (old config), not archive.db. Fixed by updating docker-compose.yml + `up -d`.
- **Video playback speed**: Browser `video.playbackRate` is capped (~16× max in Chrome) and performs poorly for very low FPS (0.1–0.3 fps) screenpipe videos. Each chunk may only be 1–2 seconds of wall-clock encoded video, playing through instantly. Solution: implement timer-based frame-jumping for speed > 2× (in progress).
- **Playhead visibility**: White 2px line was hard to see. Fixed CSS to 3px width + stronger box-shadow + added `.tt-playhead-time` overlay label style.
6. All User Messages:
- "do we use db.sqlite-shm and db.sqlite-wal for anything right now? [image of search UI] Lets add ability to search with combination... Explain how does the search work time seems to work in utc maybe display date next to the time in results. Make only apps and skip apps multiselect with all available apps."
- "doi I need to restart?"
- "there is sudo docker compose"
- "no still the same"
- "add git. I have created gitea repo https://gitea.com/lakylak/screenpipe.git"
- "just do it" (re: git push)
- "how to make sure data in :8767 are accurate and point to archive.db"
- "I can't see older than 16th there should 21st of april. In http://[IP_ADDRESS]:8766/ I can see all but not on sqlite viewer http://[IP_ADDRESS]:8767"
- "ok lets add additional port to read db.sqlite, I would like to see the structure of official db that reads db.sqlite db and one for archive"
- [docker compose output showing orphan container error]
- "looks great. I want to have ability to scroll left and right and also to jump to specific time. I want to be more aware to what I see. Still it should be fast"
- "lets make the playing speed more than 8x"
- "still playing speed is not showing properly still just in one second. also I want to see in timeline where I am"
7. Pending Tasks:
- **Playback speed fix** (in progress, partially done):
- Add `<div class="tt-playhead-time" id="ttPlayheadTime"></div>` inside `ttInner` HTML
- Implement `_ttStartSpeedTimer(speed)` / `_ttStopSpeedTimer()` / `ttSetSpeed(speed)` functions
- Update speed select `onchange` to call `ttSetSpeed(+this.value)`
- Add 64× option to speed select
- Update `_ttMovePlayhead` to also update `ttPlayheadTime` label
- Update `ttSyncBtn` to start/stop timer based on play state + speed
- Update `ttNextChunk` and `ttSeekToMs` to stop timer (restarted by onplay → ttSyncBtn)
- Rebuild Docker image after changes
8. Current Work:
Working on fixing two issues with the Timetable tab:
1. **Playback speed**: `video.playbackRate` doesn't work well for low-FPS screenpipe videos. The plan is to implement a `setInterval`-based frame-jumper for speed > 2× that advances `video.currentTime` by `speed * intervalMs / 1000` every 100–200ms, bypassing browser playbackRate limits.
2. **Timeline position indicator**: Playhead not prominent enough.
**CSS already changed** in `/volume1/screenpipe/app/static/index.html`:
```css
.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
```
Session was interrupted by summary request after this CSS edit, mid-implementation.
9. Optional Next Step:
Continuing the fix for: "still playing speed is not showing properly still just in one second. also I want to see in timeline where I am"
The next steps in order are:
1. Add `<div class="tt-playhead-time" id="ttPlayheadTime"></div>` to the HTML inside `#ttInner` (after the existing `.tt-playhead` div)
2. Update `_ttMovePlayhead` to set `ttPlayheadTime` position and text content with `fmtTime()`
3. Add timer-based speed system to JS (new functions `_ttStartSpeedTimer`, `_ttStopSpeedTimer`, `ttSetSpeed`)
4. Update speed select's `onchange` to `ttSetSpeed(+this.value)` and add 64× option
5. Update `ttSyncBtn`, `ttNextChunk`, `ttSeekToMs` to integrate with timer
6. Rebuild: `sudo docker compose build screenpipe-app && sudo docker compose up -d screenpipe-app`
If you need specific details from before compaction (like exact code snippets, error messages, or content you generated), read the full transcript at: /home/Adm1n/.claude/projects/-volume1-screenpipe/7777f46e-1e40-4051-b4bb-f43aedd59556.jsonl
Continue the conversation from where it left off without asking the user any further questions. Resume directly — do not acknowledge the summary, do not recap what was happening, do not preface with "I'll continue" or similar. Pick up the last task as if the break never happened.
Thinking
Thinking
Bash
Find timetable speed and playhead related code
IN
grep -n "tt-playhead\|ttPlayhead\|ttInner\|_ttMovePlayhead\|ttSetSpeed\|_ttStartSpeed\|_ttStopSpeed\|ttSyncBtn\|playbackRate\|speed.*select\|spdSel\|ttSpeed" /volume1/screenpipe/app/static/index.html | head -80
Copy code to clipboard
OUT
207:.tt-playhead { position:absolute; top:0; width:3px; height:36px; background:white; box-shadow:0 0 8px rgba(255,255,255,.9); z-index:10; pointer-events:none; display:none; }
208:.tt-playhead-time { position:absolute; top:2px; font-size:9px; font-weight:600; color:white; background:rgba(0,0,0,.75); padding:1px 4px; border-radius:2px; white-space:nowrap; transform:translateX(-50%); z-index:12; pointer-events:none; display:none; }
580: <div class="tt-inner" id="ttInner" onclick="ttTimelineClick(event)">
583: <div class="tt-playhead" id="ttPlayhead"></div>
598: onpause="ttSyncBtn()"
599: onplay="ttSyncBtn()">
607: <select id="ttSpeed" onchange="document.getElementById('ttVideo').playbackRate=+this.value">
1624: document.getElementById('ttInner').style.width = '100%';
1632: document.getElementById('ttPlayhead').style.display = 'none';
1694: document.getElementById('ttInner').style.width = (_ttZoom * 100) + '%';
1700: const inner = document.getElementById('ttInner');
1710: const inner = document.getElementById('ttInner');
1718: const inner = document.getElementById('ttInner');
1750: _ttMovePlayhead(new Date(best.ts).getTime());
1772: _ttMovePlayhead(new Date(best.ts).getTime());
1791:function _ttMovePlayhead(tsMs) {
1792: const ph = document.getElementById('ttPlayhead');
1802: const ph = document.getElementById('ttPlayhead');
1805: const inner = document.getElementById('ttInner');
1819:function ttSyncBtn() {
Read
index.html
index.html
(lines 576-625)
Read
index.html
index.html
(lines 1786-1865)
Thinking
Thinking
Bash
Find chunk loading and playhead update functions
IN
grep -n "ttUpdatePlayhead\|ttNextChunk\|ttLoadChunk\|_ttCurChunk\|_ttChunks\|_ttFrames" /volume1/screenpipe/app/static/index.html | head -50
Copy code to clipboard
OUT
596: onended="ttNextChunk()"
597: ontimeupdate="ttUpdatePlayhead()"
1569:let _ttFrames = [];
1570:let _ttChunks = [];
1571:let _ttCurChunk = -1;
1607: _ttFrames = data.frames;
1609: for (const f of _ttFrames) {
1613: _ttChunks = [...chunkMap.values()];
1614: _ttCurChunk = -1;
1616: _ttDayStart = new Date(_ttFrames[0].ts).getTime();
1617: _ttDayEnd = new Date(_ttFrames[_ttFrames.length - 1].ts).getTime();
1620: for (const f of _ttFrames) _ttColor(f.app);
1640: for (const f of _ttFrames) {
1736: if (!_ttFrames.length) return;
1737: let best = _ttFrames[0], minDiff = Infinity;
1738: for (const f of _ttFrames) {
1742: const ci = _ttChunks.findIndex(ch => ch[0].video_url === best.video_url);
1744: _ttCurChunk = ci;
1753:function ttNextChunk() {
1754: if (_ttCurChunk + 1 >= _ttChunks.length) return;
1755: _ttCurChunk++;
1757: video.src = _ttChunks[_ttCurChunk][0].video_url;
1762:function ttUpdatePlayhead() {
1764: if (_ttCurChunk < 0 || _ttCurChunk >= _ttChunks.length) return;
1765: const chunk = _ttChunks[_ttCurChunk];
Read
index.html
index.html
(lines 1754-1793)
Thinking
Thinking
Read
index.html
index.html
(lines 1701-1760)
Thinking
Thinking
Read
index.html
index.html
(lines 1561-1640)
Thinking
Thinking
Bash
Find time formatting functions
IN
grep -n "function fmtTime\|function fmtDate\|function _fmt\|localTimeToUtc\|fmtDateTime" /volume1/screenpipe/app/static/index.html | head -20
Copy code to clipboard
OUT
1088: if (tf) url += `&time_from=${localTimeToUtc(tf)}`;
1089: if (tt) url += `&time_to=${localTimeToUtc(tt)}`;
1153: let meta = `<span style="color:var(--muted)">${fmtDateTime(r.timestamp)}</span>${srcBadge}`;
1166: meta = `<span style="color:var(--muted)">${fmtDateTime(r.timestamp)}</span>${srcBadge} <span class="badge ${r.is_input_device?'mic':'sys'}">${r.is_input_device?'MIC':'SYS'}</span>`;
1307:function fmtTime(ts) {
1312:function fmtDateTime(ts) {
1320:function localTimeToUtc(timeStr) {
1778:function _fmtSec(s) {
1788: `${fmtDateTime(f.ts)} · ${f.app}${f.win ? ' / ' + f.win : ''}`;
Read
index.html
index.html
(lines 1308-1327)
Thinking
Thinking
Now I have everything I need. I'll implement the frame-by-frame timer system that replaces
video.playbackRate
with wall-clock accurate playback, plus add the playhead time label. Let me make the changes:
Edit
index.html
index.html
Added 1 line...
|
Code
|
screenpipe_sync.sh — screenpipe [SSH: nas]
|
NULL
|
|
2026-04-23T09:13:09.380863Z INFO sck_rs::stream_m 2026-04-23T09:13:09.380863Z INFO sck_rs::stream_manager: invalidated persistent stream for display 2
2026-04-23T09:13:32.742255Z INFO sck_rs::stream_manager: persistent SCK stream started for display 2 (3008x1253, 2fps, 3 excluded)
2026-04-23T09:14:01.764079Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=4702465002163205464, trigger=click)
2026-04-23T09:14:04.223149Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=4702465002163205464, trigger=visual_change)
2026-04-23T09:14:06.953796Z WARN sqlx::query: summary="SELECT id, snapshot_path, device_name, …" db.statement="\n\nSELECT\n id,\n snapshot_path,\n device_name,\n timestamp\nFROM\n frames\nWHERE\n snapshot_path IS NOT NULL\n AND timestamp < ?1\nORDER BY\n device_name,\n timestamp ASC\nLIMIT\n 5000\n" rows_affected=0 rows_returned=147 elapsed=6.3719465s
2026-04-23T09:14:06.954006Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 147 eligible frames
2026-04-23T09:14:12.566935Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 70 frames, 10.3MB → 2.7MB (3.8x), 70 JPEGs deleted
2026-04-23T09:14:17.358704Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 75 frames, 9.6MB → 3.0MB (3.2x), 75 JPEGs deleted
2026-04-23T09:15:47.544198Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-3660881845293713510, trigger=visual_change)
2026-04-23T09:15:50.997492Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-3660881845293713510, trigger=click)
2026-04-23T09:15:51.600725Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-3660881845293713510, trigger=click)
2026-04-23T09:15:55.723800Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-3660881845293713510, trigger=click)
2026-04-23T09:15:56.933987Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-3660881845293713510, trigger=visual_change)
2026-04-23T09:15:59.012409Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-3660881845293713510, trigger=visual_change)
2026-04-23T09:16:49.882963Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-3759145274440377841, trigger=visual_change)
2026-04-23T09:16:51.405936Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-3759145274440377841, trigger=click)
2026-04-23T09:16:52.823011Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-3759145274440377841, trigger=click)
2026-04-23T09:16:54.220304Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-3759145274440377841, trigger=click)
2026-04-23T09:16:54.887021Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-3759145274440377841, trigger=click)
2026-04-23T09:16:58.589203Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-3759145274440377841, trigger=visual_change)
2026-04-23T09:18:05.879649Z WARN screenpipe_db::db: wal checkpoint: busy (could not truncate), 5054 pages in WAL
2026-04-23T09:18:05.879619Z WARN sqlx::query: summary="PRAGMA wal_checkpoint(TRUNCATE)" db.statement="" rows_affected=0 rows_returned=1 elapsed=5.220339167s
2026-04-23T09:19:24.740122Z WARN sqlx::query: summary="SELECT id, snapshot_path, device_name, …" db.statement="\n\nSELECT\n id,\n snapshot_path,\n device_name,\n timestamp\nFROM\n frames\nWHERE\n snapshot_path IS NOT NULL\n AND timestamp < ?1\nORDER BY\n device_name,\n timestamp ASC\nLIMIT\n 5000\n" rows_affected=0 rows_returned=2 elapsed=7.381061833s
2026-04-23T09:19:24.740850Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 2 eligible frames
2026-04-23T09:19:36.514406Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=2079049776495123607, trigger=click)
2026-04-23T09:24:26.477446Z WARN sqlx::query: summary="SELECT id, snapshot_path, device_name, …" db.statement="\n\nSELECT\n id,\n snapshot_path,\n device_name,\n timestamp\nFROM\n frames\nWHERE\n snapshot_path IS NOT NULL\n AND timestamp < ?1\nORDER BY\n device_name,\n timestamp ASC\nLIMIT\n 5000\n" rows_affected=0 rows_returned=11 elapsed=1.729545666s
2026-04-23T09:24:26.477540Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 11 eligible frames
2026-04-23T09:24:27.316363Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 5 frames, 1.4MB → 1.4MB (1.0x), 5 JPEGs deleted
2026-04-23T09:24:27.966623Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 6 frames, 0.6MB → 0.2MB (2.7x), 6 JPEGs deleted
2026-04-23T09:24:42.606557Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T09:26:03.160247Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=7786631749263421517, trigger=click)
2026-04-23T09:27:26.154494Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:27:34.433056Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:28:06.713361Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:28:33.527151Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=5209591986308457575, trigger=click)
2026-04-23T09:28:34.249353Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=5209591986308457575, trigger=click)
2026-04-23T09:28:36.237199Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=5209591986308457575, trigger=click)
2026-04-23T09:28:42.253183Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=8747303508868961364, trigger=click)
2026-04-23T09:28:48.282686Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:28:50.613134Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:28:51.314522Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T09:29:31.961577Z WARN sqlx::query: summary="SELECT id, snapshot_path, device_name, …" db.statement="\n\nSELECT\n id,\n snapshot_path,\n device_name,\n timestamp\nFROM\n frames\nWHERE\n snapshot_path IS NOT NULL\n AND timestamp < ?1\nORDER BY\n device_name,\n timestamp ASC\nLIMIT\n 5000\n" rows_affected=0 rows_returned=67 elapsed=3.996031s
2026-04-23T09:29:31.961670Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 67 eligible frames
2026-04-23T09:29:34.332658Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 29 frames, 5.8MB → 2.1MB (2.8x), 29 JPEGs deleted
2026-04-23T09:29:38.223511Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 36 frames, 7.4MB → 3.8MB (2.0x), 36 JPEGs deleted
2026-04-23T09:30:34.230785Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:30:39.665086Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:30:40.299751Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T09:30:45.663893Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:30:51.622228Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:30:52.321014Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T09:31:15.021588Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T09:31:15.691601Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:31:17.946692Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T09:31:18.653547Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:32:17.960360Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:32:24.051490Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:32:29.442560Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T09:32:30.131038Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:32:32.452540Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:32:36.585069Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:32:39.322258Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:33:18.072421Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:33:18.766929Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T09:33:22.098409Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T09:33:22.782084Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:34:27.313290Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T09:34:38.542722Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 89 eligible frames
2026-04-23T09:34:41.684595Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 38 frames, 5.6MB → 0.7MB (7.5x), 38 JPEGs deleted
2026-04-23T09:34:44.933963Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 49 frames, 10.6MB → 3.8MB (2.8x), 49 JPEGs deleted
2026-04-23T09:37:57.963204Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:37:58.697604Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:38:02.124214Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:38:53.190171Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:38:59.279885Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:39:02.308536Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:39:36.189861Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:39:49.122120Z WARN sqlx::query: summary="SELECT id, snapshot_path, device_name, …" db.statement="\n\nSELECT\n id,\n snapshot_path,\n device_name,\n timestamp\nFROM\n frames\nWHERE\n snapshot_path IS NOT NULL\n AND timestamp < ?1\nORDER BY\n device_name,\n timestamp ASC\nLIMIT\n 5000\n" rows_affected=0 rows_returned=50 elapsed=4.188474542s
2026-04-23T09:39:49.122232Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 50 eligible frames
2026-04-23T09:39:51.115025Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 24 frames, 2.9MB → 0.3MB (9.4x), 24 JPEGs deleted
2026-04-23T09:39:53.096763Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 24 frames, 5.0MB → 2.3MB (2.2x), 24 JPEGs deleted
2026-04-23T09:40:57.095660Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:40:58.430714Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:41:01.130926Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T09:41:01.840278Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:41:04.072034Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T09:41:09.252541Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:41:09.972092Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T09:41:12.313228Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:41:13.696141Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T09:41:15.723102Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:41:16.452657Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:42:46.382248Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:43:42.993741Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=6469703639183038647, trigger=visual_change)
2026-04-23T09:44:17.245560Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=7072623502562616708, trigger=visual_change)
2026-04-23T09:44:56.342637Z WARN sqlx::query: summary="SELECT id, snapshot_path, device_name, …" db.statement="\n\nSELECT\n id,\n snapshot_path,\n device_name,\n timestamp\nFROM\n frames\nWHERE\n snapshot_path IS NOT NULL\n AND timestamp < ?1\nORDER BY\n device_name,\n timestamp ASC\nLIMIT\n 5000\n" rows_affected=0 rows_returned=21 elapsed=2.944921084s
2026-04-23T09:44:56.343212Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 21 eligible frames
2026-04-23T09:44:57.216675Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 10 frames, 1.2MB → 0.2MB (6.9x), 10 JPEGs deleted
2026-04-23T09:44:58.298212Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 9 frames, 2.1MB → 0.7MB (2.8x), 9 JPEGs deleted
2026-04-23T09:45:10.982752Z INFO screenpipe_engine::meeting_detector: meeting v2: Idle -> Confirming (app=Firefox, signals=2)
2026-04-23T09:45:16.376695Z INFO screenpipe_engine::meeting_detector: meeting v2: Confirming -> Active (app=Firefox, signals=2, browser=true)
2026-04-23T09:45:16.378791Z INFO screenpipe_engine::meeting_detector: meeting v2: meeting started (id=9, app=Firefox, title=None)
2026-04-23T09:50:01.989691Z WARN sqlx::query: summary="SELECT id, snapshot_path, device_name, …" db.statement="\n\nSELECT\n id,\n snapshot_path,\n device_name,\n timestamp\nFROM\n frames\nWHERE\n snapshot_path IS NOT NULL\n AND timestamp < ?1\nORDER BY\n device_name,\n timestamp ASC\nLIMIT\n 5000\n" rows_affected=0 rows_returned=23 elapsed=3.688976042s
2026-04-23T09:50:01.989962Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 23 eligible frames
2026-04-23T09:50:03.098505Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 11 frames, 1.3MB → 0.2MB (7.4x), 11 JPEGs deleted
2026-04-23T09:50:04.627741Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 10 frames, 2.4MB → 0.8MB (3.1x), 10 JPEGs deleted
2026-04-23T09:50:07.629323Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-4672421675042950776, trigger=click)
2026-04-23T09:52:11.400798Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:52:15.346773Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:52:18.174934Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:52:22.107623Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:52:32.020499Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:53:33.890239Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:53:35.304629Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T09:53:58.176084Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:54:14.248454Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:54:18.287472Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:54:20.289467Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T09:54:24.904239Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:54:27.981243Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T09:54:29.172367Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:54:30.192245Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T09:54:31.161165Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T09:55:11.821088Z WARN sqlx::query: summary="SELECT id, snapshot_path, device_name, …" db.statement="\n\nSELECT\n id,\n snapshot_path,\n device_name,\n timestamp\nFROM\n frames\nWHERE\n snapshot_path IS NOT NULL\n AND timestamp < ?1\nORDER BY\n device_name,\n timestamp ASC\nLIMIT\n 5000\n" rows_affected=0 rows_returned=36 elapsed=7.192436875s
2026-04-23T09:55:11.821248Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 36 eligible frames
2026-04-23T09:55:13.532739Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 19 frames, 2.3MB → 0.5MB (5.0x), 19 JPEGs deleted
2026-04-23T09:55:16.625152Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 15 frames, 3.3MB → 1.4MB (2.4x), 15 JPEGs deleted
2026-04-23T09:56:01.584150Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:56:24.887186Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3078857793355977422, trigger=click)
2026-04-23T09:56:32.005363Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3078857793355977422, trigger=visual_change)
2026-04-23T09:56:41.374227Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3078857793355977422, trigger=visual_change)
2026-04-23T09:56:54.666839Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3078857793355977422, trigger=click)
2026-04-23T09:57:06.486753Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3593283938725328408, trigger=visual_change)
2026-04-23T09:57:14.384178Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=5341343601501307187, trigger=visual_change)
2026-04-23T09:57:32.721765Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=5341343601501307187, trigger=visual_change)
2026-04-23T09:59:03.111441Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T09:59:18.292184Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:59:24.203906Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:59:28.763273Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T09:59:36.189282Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T10:00:16.817081Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 72 eligible frames
2026-04-23T10:00:21.637421Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 45 frames, 7.5MB → 3.5MB (2.1x), 45 JPEGs deleted
2026-04-23T10:00:24.683587Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 25 frames, 4.4MB → 1.4MB (3.2x), 25 JPEGs deleted
2026-04-23T10:00:40.573974Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-6464745230001326344, trigger=visual_change)
2026-04-23T10:02:52.495992Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-8171437999531431641, trigger=click)
2026-04-23T10:02:53.374393Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-8171437999531431641, trigger=click)
2026-04-23T10:03:37.373391Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-1961095499263201650, trigger=click)
2026-04-23T10:04:58.338820Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T10:05:04.300061Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T10:05:10.114887Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T10:05:16.104422Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T10:05:20.754499Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T10:05:25.933831Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T10:05:30.717296Z WARN sqlx::query: summary="SELECT id, snapshot_path, device_name, …" db.statement="\n\nSELECT\n id,\n snapshot_path,\n device_name,\n timestamp\nFROM\n frames\nWHERE\n snapshot_path IS NOT NULL\n AND timestamp < ?1\nORDER BY\n device_name,\n timestamp ASC\nLIMIT\n 5000\n" rows_affected=0 rows_returned=45 elapsed=5.043796209s
2026-04-23T10:05:30.717708Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 45 eligible frames
2026-04-23T10:05:33.129158Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 21 frames, 3.5MB → 1.3MB (2.8x), 21 JPEGs deleted
2026-04-23T10:05:35.719261Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 22 frames, 3.9MB → 1.6MB (2.5x), 22 JPEGs deleted
2026-04-23T10:06:54.427408Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T10:06:58.568476Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T10:07:35.294176Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T10:07:36.671700Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=click)
2026-04-23T10:07:46.281682Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=3407362332434637772, trigger=click)
2026-04-23T10:10:42.001975Z WARN sqlx::query: summary="SELECT id, snapshot_path, device_name, …" db.statement="\n\nSELECT\n id,\n snapshot_path,\n device_name,\n timestamp\nFROM\n frames\nWHERE\n snapshot_path IS NOT NULL\n AND timestamp < ?1\nORDER BY\n device_name,\n timestamp ASC\nLIMIT\n 5000\n" rows_affected=0 rows_returned=40 elapsed=6.278440792s
2026-04-23T10:10:42.002064Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 40 eligible frames
2026-04-23T10:10:45.302375Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 21 frames, 3.7MB → 1.7MB (2.2x), 21 JPEGs deleted
2026-04-23T10:10:47.110830Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 17 frames, 4.4MB → 0.6MB (7.8x), 17 JPEGs deleted
2026-04-23T10:12:18.002465Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-7144568667011834365, trigger=click)
2026-04-23T10:13:16.229637Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3362537224271461493, trigger=visual_change)
2026-04-23T10:13:20.127210Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3362537224271461493, trigger=visual_change)
2026-04-23T10:15:43.498044Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-602531024127880156, trigger=visual_change)
2026-04-23T10:15:46.873827Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-602531024127880156, trigger=visual_change)
2026-04-23T10:15:49.085137Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 32 eligible frames
2026-04-23T10:15:49.496235Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-602531024127880156, trigger=visual_change)
2026-04-23T10:15:50.519319Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 17 frames, 3.0MB → 0.9MB (3.3x), 17 JPEGs deleted
2026-04-23T10:15:51.657703Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 13 frames, 3.4MB → 0.5MB (6.7x), 13 JPEGs deleted
2026-04-23T10:15:53.186964Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-602531024127880156, trigger=visual_change)
2026-04-23T10:15:55.743993Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-602531024127880156, trigger=visual_change)
2026-04-23T10:16:10.260016Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=7730332891388508974, trigger=visual_change)
2026-04-23T10:16:19.398464Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-5862100523266843105, trigger=visual_change)
2026-04-23T10:16:23.686689Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-5862100523266843105, trigger=visual_change)
2026-04-23T10:16:27.662719Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-5862100523266843105, trigger=visual_change)
2026-04-23T10:16:34.159786Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-5862100523266843105, trigger=visual_change)
2026-04-23T10:17:00.356336Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-5862100523266843105, trigger=visual_change)
2026-04-23T10:17:22.481007Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-3265723714621232205, trigger=visual_change)
2026-04-23T10:17:29.195939Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-3265723714621232205, trigger=visual_change)
2026-04-23T10:17:47.685112Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3396265690023637464, trigger=visual_change)
2026-04-23T10:17:51.199158Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3396265690023637464, trigger=visual_change)
2026-04-23T10:17:58.199222Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3396265690023637464, trigger=visual_change)
2026-04-23T10:18:06.327777Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3396265690023637464, trigger=visual_change)
2026-04-23T10:18:13.175283Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3396265690023637464, trigger=visual_change)
2026-04-23T10:18:39.739176Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3396265690023637464, trigger=visual_change)
2026-04-23T10:19:27.720698Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3396265690023637464, trigger=visual_change)
2026-04-23T10:19:31.148096Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3396265690023637464, trigger=visual_change)
2026-04-23T10:19:35.991910Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3396265690023637464, trigger=visual_change)
2026-04-23T10:19:56.855761Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3396265690023637464, trigger=visual_change)
2026-04-23T10:20:03.521058Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3396265690023637464, trigger=visual_change)
2026-04-23T10:20:10.069690Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3396265690023637464, trigger=visual_change)
2026-04-23T10:20:18.117006Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3396265690023637464, trigger=visual_change)
2026-04-23T10:20:54.473088Z WARN sqlx::query: summary="SELECT id, snapshot_path, device_name, …" db.statement="\n\nSELECT\n id,\n snapshot_path,\n device_name,\n timestamp\nFROM\n frames\nWHERE\n snapshot_path IS NOT NULL\n AND timestamp < ?1\nORDER BY\n device_name,\n timestamp ASC\nLIMIT\n 5000\n" rows_affected=0 rows_returned=27 elapsed=2.810096209s
2026-04-23T10:20:54.473356Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 27 eligible frames
2026-04-23T10:20:56.920972Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 14 frames, 2.2MB → 0.9MB (2.6x), 14 JPEGs deleted
2026-04-23T10:20:58.678464Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 11 frames, 2.5MB → 0.8MB (3.0x), 11 JPEGs deleted
2026-04-23T10:23:10.727426Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3844820439140861821, trigger=visual_change)
2026-04-23T10:25:16.468416Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T10:25:32.507777Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3407362332434637772, trigger=visual_change)
2026-04-23T10:26:05.910946Z WARN sqlx::query: summary="SELECT id, snapshot_path, device_name, …" db.statement="\n\nSELECT\n id,\n snapshot_path,\n device_name,\n timestamp\nFROM\n frames\nWHERE\n snapshot_path IS NOT NULL\n AND timestamp < ?1\nORDER BY\n device_name,\n timestamp ASC\nLIMIT\n 5000\n" rows_affected=0 rows_returned=33 elapsed=7.132200208s
2026-04-23T10:26:05.912569Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 33 eligible frames
2026-04-23T10:26:09.921886Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 18 frames, 3.0MB → 1.2MB (2.5x), 18 JPEGs deleted
2026-04-23T10:26:13.670275Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 13 frames, 2.6MB → 0.9MB (3.0x), 13 JPEGs deleted
2026-04-23T10:26:20.034712Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=click)
2026-04-23T10:26:22.436991Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=click)
2026-04-23T10:26:25.816727Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:27:46.056157Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-1615606372458852745, trigger=click)
2026-04-23T10:28:05.117587Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:28:16.307383Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:28:22.684004Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:28:39.022530Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:28:44.950116Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:28:50.090325Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:28:52.144212Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:28:55.229550Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-2090548142966872323, trigger=click)
2026-04-23T10:28:59.714370Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-2090548142966872323, trigger=click)
2026-04-23T10:29:08.010841Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-2090548142966872323, trigger=click)
2026-04-23T10:29:11.877274Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-2090548142966872323, trigger=click)
2026-04-23T10:29:15.515889Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=click)
2026-04-23T10:29:19.052095Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=click)
2026-04-23T10:29:21.109938Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=click)
2026-04-23T10:29:28.551159Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:29:30.947483Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:29:38.350095Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:29:51.477755Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:29:56.846851Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:30:02.584960Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:30:06.779849Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:30:18.311444Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:30:22.665242Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:30:29.333028Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:30:37.983139Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:30:48.829636Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-2090548142966872323, trigger=click)
2026-04-23T10:30:59.474314Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3712905987097609358, trigger=visual_change)
2026-04-23T10:31:16.115641Z WARN sqlx::query: summary="SELECT id, snapshot_path, device_name, …" db.statement="\n\nSELECT\n id,\n snapshot_path,\n device_name,\n timestamp\nFROM\n frames\nWHERE\n snapshot_path IS NOT NULL\n AND timestamp < ?1\nORDER BY\n device_name,\n timestamp ASC\nLIMIT\n 5000\n" rows_affected=0 rows_returned=34 elapsed=1.523826083s
2026-04-23T10:31:16.115774Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 34 eligible frames
2026-04-23T10:31:19.674413Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 19 frames, 3.4MB → 2.0MB (1.7x), 19 JPEGs deleted
2026-04-23T10:31:20.281876Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:31:21.222414Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 13 frames, 3.4MB → 0.8MB (4.5x), 13 JPEGs deleted
2026-04-23T10:31:30.589390Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:31:31.649068Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:31:34.542890Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:31:35.913592Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:31:42.222598Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:31:46.314096Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:31:56.457995Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=visual_change)
2026-04-23T10:32:02.440693Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-2090548142966872323, trigger=click)
2026-04-23T10:32:11.844302Z INFO screenpipe_engine::meeting_detector: meeting v2: Active -> Ending (no controls, app=Firefox, id=9, grace=300s)
2026-04-23T10:36:21.778297Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 34 eligible frames
2026-04-23T10:36:23.408803Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 19 frames, 3.2MB → 1.6MB (2.1x), 19 JPEGs deleted
2026-04-23T10:36:24.427188Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 13 frames, 3.4MB → 0.7MB (4.8x), 13 JPEGs deleted
2026-04-23T10:37:15.245266Z INFO screenpipe_engine::meeting_detector: meeting v2: Ending -> Idle (timeout=300s, app=Firefox, id=9)
2026-04-23T10:37:15.499978Z INFO screenpipe_engine::meeting_detector: meeting v2: meeting ended (id=9)
2026-04-23T10:39:08.136849Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=7072094695368023128, trigger=visual_change)
2026-04-23T10:40:10.173692Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-5626541066854626210, trigger=visual_change)
2026-04-23T10:41:03.187335Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-6908211675272154683, trigger=visual_change)
2026-04-23T10:41:27.801144Z WARN sqlx::query: summary="SELECT id, snapshot_path, device_name, …" db.statement="\n\nSELECT\n id,\n snapshot_path,\n device_name,\n timestamp\nFROM\n frames\nWHERE\n snapshot_path IS NOT NULL\n AND timestamp < ?1\nORDER BY\n device_name,\n timestamp ASC\nLIMIT\n 5000\n" rows_affected=0 rows_returned=32 elapsed=3.230263875s
2026-04-23T10:41:27.801229Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 32 eligible frames
2026-04-23T10:41:30.240257Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 17 frames, 2.7MB → 1.1MB (2.4x), 17 JPEGs deleted
2026-04-23T10:41:31.414104Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 13 frames, 3.3MB → 1.1MB (3.1x), 13 JPEGs deleted
2026-04-23T10:43:59.795473Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=5090325678268650097, trigger=click)
2026-04-23T10:44:00.574274Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=5090325678268650097, trigger=click)
2026-04-23T10:44:03.353223Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=5090325678268650097, trigger=visual_change)
2026-04-23T10:44:06.333413Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=5090325678268650097, trigger=visual_change)
2026-04-23T10:44:15.554031Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=5090325678268650097, trigger=click)
2026-04-23T10:44:16.272856Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=5090325678268650097, trigger=click)
2026-04-23T10:44:36.868831Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=5090325678268650097, trigger=visual_change)
2026-04-23T10:44:52.509317Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=5090325678268650097, trigger=visual_change)
2026-04-23T10:45:05.711036Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=5090325678268650097, trigger=visual_change)
2026-04-23T10:45:15.149681Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=5090325678268650097, trigger=click)
2026-04-23T10:45:15.874930Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=5090325678268650097, trigger=click)
2026-04-23T10:45:18.722067Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=5090325678268650097, trigger=click)
2026-04-23T10:45:27.888960Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=5090325678268650097, trigger=visual_change)
2026-04-23T10:45:32.102259Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=5090325678268650097, trigger=click)
2026-04-23T10:45:47.964284Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-250182447663358748, trigger=click)
2026-04-23T10:45:57.352178Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-250182447663358748, trigger=click)
2026-04-23T10:45:58.059330Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-250182447663358748, trigger=click)
2026-04-23T10:46:00.786122Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-250182447663358748, trigger=click)
2026-04-23T10:46:01.515587Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=-250182447663358748, trigger=click)
2026-04-23T10:46:32.335524Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 34 eligible frames
2026-04-23T10:46:33.454412Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 15 frames, 1.3MB → 0.3MB (4.3x), 15 JPEGs deleted
2026-04-23T10:46:35.185172Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 17 frames, 4.4MB → 0.8MB (5.1x), 17 JPEGs deleted
2026-04-23T10:49:51.424053Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=5090325678268650097, trigger=click)
2026-04-23T10:49:52.278022Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=5090325678268650097, trigger=click)
2026-04-23T10:50:06.945403Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=5090325678268650097, trigger=visual_change)
2026-04-23T10:50:20.071306Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=5090325678268650097, trigger=click)
2026-04-23T10:50:27.322330Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 2 (hash=5090325678268650097, trigger=click)
2026-04-23T10:50:28.025501Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping captur...
|
iTerm2
|
screenpipe" — 95✕45
|
NULL
|
|
Last login: Thu Apr 9 19:52:08 on ttys007
Poetry Last login: Thu Apr 9 19:52:08 on ttys007
Poetry could not find a pyproject.toml file in /Users/lukas/jiminny/app or its parents
Poetry could not find a pyproject.toml file in /Users/lukas/jiminny/app or its parents
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/jiminny/app (JY-18909-automated-reports-ask-jiminny) $ cd
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record
detected hardware tier: Mid
checking permissions...
screen recording: ok
microphone: ok
accessibility: ok
2026-04-09T19:53:07.212588Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T19:53:08.176294Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T19:53:08.180179Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T19:53:08.180172Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T19:53:08.180644Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T19:53:08.181079Z INFO screenpipe_audio::speaker::models: initiating wespeaker_en_voxceleb_CAM++.onnx model download...
2026-04-09T19:53:08.181124Z WARN screenpipe_audio::segmentation::segmentation_manager: embedding model unavailable at startup: wespeaker_en_voxceleb_CAM++.onnx model not available yet; download started in background
2026-04-09T19:53:08.181175Z INFO screenpipe_audio::speaker::models: initiating segmentation-3.0.onnx model download...
2026-04-09T19:53:08.181183Z WARN screenpipe_audio::segmentation::segmentation_manager: segmentation model unavailable at startup: segmentation-3.0.onnx model not available yet; download started in background
2026-04-09T19:53:08.181157Z INFO screenpipe_audio::speaker::models: wespeaker_en_voxceleb_CAM++.onnx model download attempt 1/3
2026-04-09T19:53:08.181223Z WARN screenpipe_audio::audio_manager::manager: silero vad unavailable, falling back to webrtc: silero vad model not available yet; download started in background
2026-04-09T19:53:08.181225Z INFO screenpipe_audio::speaker::models: downloading wespeaker_en_voxceleb_CAM++.onnx model from https://github.com/screenpipe/screenpipe/raw/refs/heads/main/crates/screenpipe-audio/models/pyannote/wespeaker_en_voxceleb_CAM++.onnx
2026-04-09T19:53:08.181212Z INFO screenpipe_audio::speaker::models: segmentation-3.0.onnx model download attempt 1/3
2026-04-09T19:53:08.181526Z INFO screenpipe_audio::speaker::models: downloading segmentation-3.0.onnx model from https://github.com/screenpipe/screenpipe/raw/refs/heads/main/crates/screenpipe-audio/models/pyannote/segmentation-3.0.onnx
2026-04-09T19:53:08.181252Z INFO screenpipe_audio::vad::silero: downloading Silero VAD v5 model...
2026-04-09T19:53:08.181607Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T19:53:08.181646Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T19:53:08.181366Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T19:53:08.181759Z INFO screenpipe_core::pipes: installed built-in pipe: day-recap
2026-04-09T19:53:08.181952Z INFO screenpipe_core::pipes: installed built-in pipe: standup-update
2026-04-09T19:53:08.182315Z INFO screenpipe_core::pipes: installed built-in pipe: ai-habits
2026-04-09T19:53:08.182501Z INFO screenpipe_core::pipes: installed built-in pipe: time-breakdown
2026-04-09T19:53:08.182680Z INFO screenpipe_core::pipes: installed built-in pipe: video-export
2026-04-09T19:53:08.182875Z INFO screenpipe_core::pipes: installed built-in pipe: meeting-summary
2026-04-09T19:53:08.183091Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T19:53:08.183146Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T19:53:08.183187Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T19:53:08.183237Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T19:53:08.183306Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T19:53:08.183363Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T19:53:08.183380Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ false │
│ vision disabled │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ soundcore AeroClip (input) │
│ │ System Audio (output) │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: https://github.com/screenpipe/screenpipe/releases
2026-04-09T19:53:08.185865Z INFO screenpipe: starting UI event capture
2026-04-09T19:53:08.185586Z INFO screenpipe_core::pipes: pipe scheduler started (generation 2)
2026-04-09T19:53:08.188922Z WARN screenpipe: pi agent install failed: bun not found — install from https://bun.sh
2026-04-09T19:53:08.194461Z INFO screenpipe_engine::power::manager: initial power profile: Performance (on_ac=true, battery=Some(100))
2026-04-09T19:53:08.201212Z INFO screenpipe_engine::ui_recorder: Starting UI event capture
2026-04-09T19:53:08.217633Z INFO screenpipe_engine::calendar_speaker_id: speaker identification: started (user_name=<not set>)
2026-04-09T19:53:08.217794Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warming from DB (2026-04-08 16:53:08.217792 UTC to 2026-04-09 16:53:08.217792 UTC)
2026-04-09T19:53:08.217649Z INFO screenpipe_engine::ui_recorder: UI recording session started: 3b59b1d1-d80e-4689-8c72-6476af4571df
2026-04-09T19:53:08.220368Z INFO screenpipe_engine::meeting_detector: meeting v2: detection loop started (base_interval=5s, profiles=12)
2026-04-09T19:53:08.222489Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warmed with 0 frame entries, coverage from 2026-04-08 16:53:08.217792 UTC
2026-04-09T19:53:08.227257Z INFO screenpipe_engine::server: Server listening on [IP_ADDRESS]:3030
2026-04-09T19:53:08.232250Z INFO screenpipe_connect::mdns: mdns: advertising screenpipe on port 3030
2026-04-09T19:53:08.543211Z INFO screenpipe_engine::vision_manager::manager: Starting vision recording for monitor 1 (1440x900)
2026-04-09T19:53:08.543241Z INFO screenpipe_engine::vision_manager::manager: Starting event-driven capture for monitor 1 (device: monitor_1)
2026-04-09T19:53:08.543253Z INFO screenpipe_engine::vision_manager::manager: Skipping monitor 2 (Display 2_3008x1253_-813,-1253) — not in allowed list
2026-04-09T19:53:08.543262Z INFO screenpipe_engine::vision_manager::monitor_watcher: Starting monitor watcher (polling every 5 seconds)
2026-04-09T19:53:08.543299Z INFO screenpipe_engine::event_driven_capture: event-driven capture started for monitor 1 (device: monitor_1)
2026-04-09T19:53:09.210453Z INFO screenpipe_audio::vad::silero: Silero VAD v5 model downloaded and saved to: "/Users/lukas/Library/Caches/screenpipe/vad/silero_vad_v5.onnx"
2026-04-09T19:53:09.352790Z INFO screenpipe_audio::speaker::models: saving segmentation-3.0.onnx model (5983836 bytes) to "/Users/lukas/Library/Caches/screenpipe/models/segmentation-3.0.onnx"
2026-04-09T19:53:09.355799Z INFO screenpipe_audio::speaker::models: segmentation-3.0.onnx model successfully downloaded and saved
2026-04-09T19:53:09.367746Z INFO sck_rs::stream_manager: persistent SCK stream started for display 1 (1440x900, 2fps)
2026-04-09T19:53:10.102366Z INFO screenpipe_audio::speaker::models: saving wespeaker_en_voxceleb_CAM++.onnx model (29292684 bytes) to "/Users/lukas/Library/Caches/screenpipe/models/wespeaker_en_voxceleb_CAM++.onnx"
2026-04-09T19:53:10.112264Z INFO screenpipe_audio::speaker::models: wespeaker_en_voxceleb_CAM++.onnx model successfully downloaded and saved
2026-04-09T19:53:13.221908Z INFO screenpipe_engine::retention: retention: initialized with 14d retention
2026-04-09T19:53:13.222855Z INFO screenpipe: local retention auto-enabled (14 days)
2026-04-09T19:53:15.841867Z INFO screenpipe_engine::event_driven_capture: startup capture for monitor 1: frame_id=1, dur=797ms
2026-04-09T19:53:18.189478Z WARN screenpipe_audio::transcription::engine: whisper model is not available locally yet for WhisperTiny; audio transcription disabled until download completes
2026-04-09T19:53:18.189794Z INFO screenpipe_audio::audio_manager::manager: transcription session created (will be reused across segments)
2026-04-09T19:53:18.190227Z INFO screenpipe_audio::transcription::whisper::model: downloading model "ggml-tiny.bin"
2026-04-09T19:53:18.190847Z INFO screenpipe_audio::audio_manager::manager: audio manager started
2026-04-09T19:53:18.190916Z INFO screenpipe_audio::audio_manager::manager: calendar-assisted speaker diarization: listening for meeting events
ggml-tiny.bin [00:00:01] [███████████████████████████████████████████████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] 38.34 MiB/74.09 MiB 21.65 MiB/s (2s)2026-04-09T19:53:19.720503Z INFO screenpipe_audio::audio_manager::device_monitor: model capabilities changed after background download; restarting central handlers
2026-04-09T19:53:19.720675Z WARN screenpipe_audio::transcription::engine: whisper model is not available locally yet for WhisperTiny; audio transcription disabled until download completes
2026-04-09T19:53:19.720732Z INFO screenpipe_audio::audio_manager::manager: transcription session created (will be reused across segments)
2026-04-09T19:53:19.720787Z INFO screenpipe_audio::audio_manager::manager: central audio-receiver handler restarted for capability refresh
2026-04-09T19:53:19.720795Z INFO screenpipe_audio::audio_manager::manager: central transcription-receiver handler restarted for capability refresh
2026-04-09T19:53:19.721204Z INFO screenpipe_audio::transcription::whisper::model: downloading model "ggml-tiny.bin"
ggml-tiny.bin [00:00:01] [██████████████████████████████████████████████████████████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] 45.66 MiB/74.09 MiB 23.13 MiB/s (1s)
ggml-tiny.bin [00:00:00] [███░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] 2.09 MiB/74.09 MiB 1.63 MiB/s (44s)2026-04-09T19:53:20.383627Z INFO screenpipe_audio::device::device_manager: starting recording for device: System Audio (output)
2026-04-09T19:53:20.383837Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
ggml-tiny.bin [00:00:01] [██████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] 9.22 MiB/74.09 MiB 4.95 MiB/s (13s)2026-04-09T19:53:20.972014Z INFO screenpipe_audio::device::device_manager: starting recording for device: soundcore AeroClip (input)
2026-04-09T19:53:20.972101Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for soundcore AeroClip (input) (30s segments)
ggml-tiny.bin [00:00:02] [███████████████████████████████████████████████████████████████████████████████████████████████████████████████████] 74.09 MiB/74.09 MiB 28.28 MiB/s (0s)2026-04-09T19:53:21.047457Z INFO screenpipe_audio::transcription::whisper::model: model downloaded ggml-tiny.bin
2026-04-09T19:53:21.047589Z INFO screenpipe_audio::transcription::engine: whisper model downloaded in background: "/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin"
ggml-tiny.bin [00:00:03] [███████████████████████████████████████████████████████████████████████████████████████████████████████████████████] 74.09 MiB/74.09 MiB 23.56 MiB/s (0s)2026-04-09T19:53:23.080110Z INFO screenpipe_audio::transcription::whisper::model: model downloaded ggml-tiny.bin
2026-04-09T19:53:23.080240Z INFO screenpipe_audio::transcription::engine: whisper model downloaded in background: "/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin"
2026-04-09T19:53:52.431282Z INFO screenpipe_audio::transcription::engine: whisper model available: "/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin"
2026-04-09T19:53:52.431325Z INFO screenpipe_audio::transcription::whisper::model: whisper context: gpu acceleration enabled (Metal on macOS, Vulkan on Windows)
2026-04-09T19:53:52.431330Z INFO screenpipe_audio::transcription::engine: loading whisper model with GPU acceleration...
whisper_init_from_file_with_params_no_state: loading model from '/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin'
whisper_init_with_params_no_state: use gpu = 1
whisper_init_with_params_no_state: flash attn = 0
whisper_init_with_params_no_state: gpu_device = 0
whisper_init_with_params_no_state: dtw = 0
ggml_metal_device_init: tensor API disabled for pre-M5 and pre-A19 devices
ggml_metal_library_init: using embedded metal library
ggml_metal_library_init: loaded in 14.652 sec
ggml_metal_rsets_init: creating a residency set collection (keep_alive = 180 s)
ggml_metal_device_init: GPU name: Apple M1
ggml_metal_device_init: GPU family: MTLGPUFamilyApple7 (1007)
ggml_metal_device_init: GPU family: MTLGPUFamilyCommon3 (3003)
ggml_metal_device_init: GPU family: MTLGPUFamilyMetal3 (5001)
ggml_metal_device_init: simdgroup reduction = true
ggml_metal_device_init: simdgroup matrix mul. = true
ggml_metal_device_init: has unified memory = true
ggml_metal_device_init: has bfloat = true
ggml_metal_device_init: has tensor = false
ggml_metal_device_init: use residency sets = true
ggml_metal_device_init: use shared buffers = true
ggml_metal_device_init: recommendedMaxWorkingSetSize = 11453.25 MB
whisper_init_with_params_no_state: devices = 3
whisper_init_with_params_no_state: backends = 3
whisper_model_load: loading model
whisper_model_load: n_vocab = 51865
whisper_model_load: n_audio_ctx = 1500
whisper_model_load: n_audio_state = 384
whisper_model_load: n_audio_head = 6
whisper_model_load: n_audio_layer = 4
whisper_model_load: n_text_ctx = 448
whisper_model_load: n_text_state = 384
whisper_model_load: n_text_head = 6
whisper_model_load: n_text_layer = 4
whisper_model_load: n_mels = 80
whisper_model_load: ftype = 1
whisper_model_load: qntvr = 0
whisper_model_load: type = 1 (tiny)
whisper_model_load: adding 1608 extra tokens
whisper_model_load: n_langs = 99
whisper_model_load: Metal total size = 77.11 MB
whisper_model_load: model size = 77.11 MB
2026-04-09T19:54:07.178366Z INFO screenpipe_audio::transcription::engine: whisper model loaded successfully
2026-04-09T19:54:07.178441Z INFO screenpipe_audio::audio_manager::device_monitor: model capabilities changed after background download; restarting central handlers
2026-04-09T19:54:07.178636Z INFO screenpipe_audio::transcription::engine: whisper model available: "/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin"
2026-04-09T19:54:07.178655Z INFO screenpipe_audio::transcription::whisper::model: whisper context: gpu acceleration enabled (Metal on macOS, Vulkan on Windows)
2026-04-09T19:54:07.178661Z INFO screenpipe_audio::transcription::engine: loading whisper model with GPU acceleration...
whisper_init_from_file_with_params_no_state: loading model from '/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin'
whisper_init_with_params_no_state: use gpu = 1
whisper_init_with_params_no_state: flash attn = 0
whisper_init_with_params_no_state: gpu_device = 0
whisper_init_with_params_no_state: dtw = 0
whisper_init_with_params_no_state: devices = 3
whisper_init_with_params_no_state: backends = 3
whisper_model_load: loading model
whisper_model_load: n_vocab = 51865
whisper_model_load: n_audio_ctx = 1500
whisper_model_load: n_audio_state = 384
whisper_model_load: n_audio_head = 6
whisper_model_load: n_audio_layer = 4
whisper_model_load: n_text_ctx = 448
whisper_model_load: n_text_state = 384
whisper_model_load: n_text_head = 6
whisper_model_load: n_text_layer = 4
whisper_model_load: n_mels = 80
whisper_model_load: ftype = 1
whisper_model_load: qntvr = 0
whisper_model_load: type = 1 (tiny)
whisper_model_load: adding 1608 extra tokens
whisper_model_load: n_langs = 99
whisper_model_load: Metal total size = 77.11 MB
whisper_model_load: model size = 77.11 MB
2026-04-09T19:54:07.242116Z INFO screenpipe_audio::transcription::engine: whisper model loaded successfully
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
2026-04-09T19:54:07.250010Z INFO screenpipe_audio::audio_manager::manager: transcription session created (will be reused across segments)
2026-04-09T19:54:07.250055Z INFO screenpipe_audio::audio_manager::manager: central audio-receiver handler restarted for capability refresh
2026-04-09T19:54:07.250066Z INFO screenpipe_audio::audio_manager::manager: central transcription-receiver handler restarted for capability refresh
DOCKER
Close Tab
DEV (-zsh)
Close Tab
APP (-zsh)
Close Tab
screenpipe"
Close Tab
⌥⌘1
screenpipe"...
|
iTerm2
|
screenpipe"
|
NULL
|
|
Last login: Thu Apr 9 19:52:08 on ttys007
Poetry Last login: Thu Apr 9 19:52:08 on ttys007
Poetry could not find a pyproject.toml file in /Users/lukas/jiminny/app or its parents
Poetry could not find a pyproject.toml file in /Users/lukas/jiminny/app or its parents
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/jiminny/app (JY-18909-automated-reports-ask-jiminny) $ cd
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record
detected hardware tier: Mid
checking permissions...
screen recording: ok
microphone: ok
accessibility: ok
2026-04-09T19:53:07.212588Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T19:53:08.176294Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T19:53:08.180179Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T19:53:08.180172Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T19:53:08.180644Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T19:53:08.181079Z INFO screenpipe_audio::speaker::models: initiating wespeaker_en_voxceleb_CAM++.onnx model download...
2026-04-09T19:53:08.181124Z WARN screenpipe_audio::segmentation::segmentation_manager: embedding model unavailable at startup: wespeaker_en_voxceleb_CAM++.onnx model not available yet; download started in background
2026-04-09T19:53:08.181175Z INFO screenpipe_audio::speaker::models: initiating segmentation-3.0.onnx model download...
2026-04-09T19:53:08.181183Z WARN screenpipe_audio::segmentation::segmentation_manager: segmentation model unavailable at startup: segmentation-3.0.onnx model not available yet; download started in background
2026-04-09T19:53:08.181157Z INFO screenpipe_audio::speaker::models: wespeaker_en_voxceleb_CAM++.onnx model download attempt 1/3
2026-04-09T19:53:08.181223Z WARN screenpipe_audio::audio_manager::manager: silero vad unavailable, falling back to webrtc: silero vad model not available yet; download started in background
2026-04-09T19:53:08.181225Z INFO screenpipe_audio::speaker::models: downloading wespeaker_en_voxceleb_CAM++.onnx model from https://github.com/screenpipe/screenpipe/raw/refs/heads/main/crates/screenpipe-audio/models/pyannote/wespeaker_en_voxceleb_CAM++.onnx
2026-04-09T19:53:08.181212Z INFO screenpipe_audio::speaker::models: segmentation-3.0.onnx model download attempt 1/3
2026-04-09T19:53:08.181526Z INFO screenpipe_audio::speaker::models: downloading segmentation-3.0.onnx model from https://github.com/screenpipe/screenpipe/raw/refs/heads/main/crates/screenpipe-audio/models/pyannote/segmentation-3.0.onnx
2026-04-09T19:53:08.181252Z INFO screenpipe_audio::vad::silero: downloading Silero VAD v5 model...
2026-04-09T19:53:08.181607Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T19:53:08.181646Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T19:53:08.181366Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T19:53:08.181759Z INFO screenpipe_core::pipes: installed built-in pipe: day-recap
2026-04-09T19:53:08.181952Z INFO screenpipe_core::pipes: installed built-in pipe: standup-update
2026-04-09T19:53:08.182315Z INFO screenpipe_core::pipes: installed built-in pipe: ai-habits
2026-04-09T19:53:08.182501Z INFO screenpipe_core::pipes: installed built-in pipe: time-breakdown
2026-04-09T19:53:08.182680Z INFO screenpipe_core::pipes: installed built-in pipe: video-export
2026-04-09T19:53:08.182875Z INFO screenpipe_core::pipes: installed built-in pipe: meeting-summary
2026-04-09T19:53:08.183091Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T19:53:08.183146Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T19:53:08.183187Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T19:53:08.183237Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T19:53:08.183306Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T19:53:08.183363Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T19:53:08.183380Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ false │
│ vision disabled │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ soundcore AeroClip (input) │
│ │ System Audio (output) │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: https://github.com/screenpipe/screenpipe/releases
2026-04-09T19:53:08.185865Z INFO screenpipe: starting UI event capture
2026-04-09T19:53:08.185586Z INFO screenpipe_core::pipes: pipe scheduler started (generation 2)
2026-04-09T19:53:08.188922Z WARN screenpipe: pi agent install failed: bun not found — install from https://bun.sh
2026-04-09T19:53:08.194461Z INFO screenpipe_engine::power::manager: initial power profile: Performance (on_ac=true, battery=Some(100))
2026-04-09T19:53:08.201212Z INFO screenpipe_engine::ui_recorder: Starting UI event capture
2026-04-09T19:53:08.217633Z INFO screenpipe_engine::calendar_speaker_id: speaker identification: started (user_name=<not set>)
2026-04-09T19:53:08.217794Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warming from DB (2026-04-08 16:53:08.217792 UTC to 2026-04-09 16:53:08.217792 UTC)
2026-04-09T19:53:08.217649Z INFO screenpipe_engine::ui_recorder: UI recording session started: 3b59b1d1-d80e-4689-8c72-6476af4571df
2026-04-09T19:53:08.220368Z INFO screenpipe_engine::meeting_detector: meeting v2: detection loop started (base_interval=5s, profiles=12)
2026-04-09T19:53:08.222489Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warmed with 0 frame entries, coverage from 2026-04-08 16:53:08.217792 UTC
2026-04-09T19:53:08.227257Z INFO screenpipe_engine::server: Server listening on [IP_ADDRESS]:3030
2026-04-09T19:53:08.232250Z INFO screenpipe_connect::mdns: mdns: advertising screenpipe on port 3030
2026-04-09T19:53:08.543211Z INFO screenpipe_engine::vision_manager::manager: Starting vision recording for monitor 1 (1440x900)
2026-04-09T19:53:08.543241Z INFO screenpipe_engine::vision_manager::manager: Starting event-driven capture for monitor 1 (device: monitor_1)
2026-04-09T19:53:08.543253Z INFO screenpipe_engine::vision_manager::manager: Skipping monitor 2 (Display 2_3008x1253_-813,-1253) — not in allowed list
2026-04-09T19:53:08.543262Z INFO screenpipe_engine::vision_manager::monitor_watcher: Starting monitor watcher (polling every 5 seconds)
2026-04-09T19:53:08.543299Z INFO screenpipe_engine::event_driven_capture: event-driven capture started for monitor 1 (device: monitor_1)
2026-04-09T19:53:09.210453Z INFO screenpipe_audio::vad::silero: Silero VAD v5 model downloaded and saved to: "/Users/lukas/Library/Caches/screenpipe/vad/silero_vad_v5.onnx"
2026-04-09T19:53:09.352790Z INFO screenpipe_audio::speaker::models: saving segmentation-3.0.onnx model (5983836 bytes) to "/Users/lukas/Library/Caches/screenpipe/models/segmentation-3.0.onnx"
2026-04-09T19:53:09.355799Z INFO screenpipe_audio::speaker::models: segmentation-3.0.onnx model successfully downloaded and saved
2026-04-09T19:53:09.367746Z INFO sck_rs::stream_manager: persistent SCK stream started for display 1 (1440x900, 2fps)
2026-04-09T19:53:10.102366Z INFO screenpipe_audio::speaker::models: saving wespeaker_en_voxceleb_CAM++.onnx model (29292684 bytes) to "/Users/lukas/Library/Caches/screenpipe/models/wespeaker_en_voxceleb_CAM++.onnx"
2026-04-09T19:53:10.112264Z INFO screenpipe_audio::speaker::models: wespeaker_en_voxceleb_CAM++.onnx model successfully downloaded and saved
2026-04-09T19:53:13.221908Z INFO screenpipe_engine::retention: retention: initialized with 14d retention
2026-04-09T19:53:13.222855Z INFO screenpipe: local retention auto-enabled (14 days)
2026-04-09T19:53:15.841867Z INFO screenpipe_engine::event_driven_capture: startup capture for monitor 1: frame_id=1, dur=797ms
2026-04-09T19:53:18.189478Z WARN screenpipe_audio::transcription::engine: whisper model is not available locally yet for WhisperTiny; audio transcription disabled until download completes
2026-04-09T19:53:18.189794Z INFO screenpipe_audio::audio_manager::manager: transcription session created (will be reused across segments)
2026-04-09T19:53:18.190227Z INFO screenpipe_audio::transcription::whisper::model: downloading model "ggml-tiny.bin"
2026-04-09T19:53:18.190847Z INFO screenpipe_audio::audio_manager::manager: audio manager started
2026-04-09T19:53:18.190916Z INFO screenpipe_audio::audio_manager::manager: calendar-assisted speaker diarization: listening for meeting events
ggml-tiny.bin [00:00:01] [███████████████████████████████████████████████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] 38.34 MiB/74.09 MiB 21.65 MiB/s (2s)2026-04-09T19:53:19.720503Z INFO screenpipe_audio::audio_manager::device_monitor: model capabilities changed after background download; restarting central handlers
2026-04-09T19:53:19.720675Z WARN screenpipe_audio::transcription::engine: whisper model is not available locally yet for WhisperTiny; audio transcription disabled until download completes
2026-04-09T19:53:19.720732Z INFO screenpipe_audio::audio_manager::manager: transcription session created (will be reused across segments)
2026-04-09T19:53:19.720787Z INFO screenpipe_audio::audio_manager::manager: central audio-receiver handler restarted for capability refresh
2026-04-09T19:53:19.720795Z INFO screenpipe_audio::audio_manager::manager: central transcription-receiver handler restarted for capability refresh
2026-04-09T19:53:19.721204Z INFO screenpipe_audio::transcription::whisper::model: downloading model "ggml-tiny.bin"
ggml-tiny.bin [00:00:01] [██████████████████████████████████████████████████████████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] 45.66 MiB/74.09 MiB 23.13 MiB/s (1s)
ggml-tiny.bin [00:00:00] [███░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] 2.09 MiB/74.09 MiB 1.63 MiB/s (44s)2026-04-09T19:53:20.383627Z INFO screenpipe_audio::device::device_manager: starting recording for device: System Audio (output)
2026-04-09T19:53:20.383837Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
ggml-tiny.bin [00:00:01] [██████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] 9.22 MiB/74.09 MiB 4.95 MiB/s (13s)2026-04-09T19:53:20.972014Z INFO screenpipe_audio::device::device_manager: starting recording for device: soundcore AeroClip (input)
2026-04-09T19:53:20.972101Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for soundcore AeroClip (input) (30s segments)
ggml-tiny.bin [00:00:02] [███████████████████████████████████████████████████████████████████████████████████████████████████████████████████] 74.09 MiB/74.09 MiB 28.28 MiB/s (0s)2026-04-09T19:53:21.047457Z INFO screenpipe_audio::transcription::whisper::model: model downloaded ggml-tiny.bin
2026-04-09T19:53:21.047589Z INFO screenpipe_audio::transcription::engine: whisper model downloaded in background: "/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin"
ggml-tiny.bin [00:00:03] [███████████████████████████████████████████████████████████████████████████████████████████████████████████████████] 74.09 MiB/74.09 MiB 23.56 MiB/s (0s)2026-04-09T19:53:23.080110Z INFO screenpipe_audio::transcription::whisper::model: model downloaded ggml-tiny.bin
2026-04-09T19:53:23.080240Z INFO screenpipe_audio::transcription::engine: whisper model downloaded in background: "/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin"
2026-04-09T19:53:52.431282Z INFO screenpipe_audio::transcription::engine: whisper model available: "/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin"
2026-04-09T19:53:52.431325Z INFO screenpipe_audio::transcription::whisper::model: whisper context: gpu acceleration enabled (Metal on macOS, Vulkan on Windows)
2026-04-09T19:53:52.431330Z INFO screenpipe_audio::transcription::engine: loading whisper model with GPU acceleration...
whisper_init_from_file_with_params_no_state: loading model from '/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin'
whisper_init_with_params_no_state: use gpu = 1
whisper_init_with_params_no_state: flash attn = 0
whisper_init_with_params_no_state: gpu_device = 0
whisper_init_with_params_no_state: dtw = 0
ggml_metal_device_init: tensor API disabled for pre-M5 and pre-A19 devices
ggml_metal_library_init: using embedded metal library
ggml_metal_library_init: loaded in 14.652 sec
ggml_metal_rsets_init: creating a residency set collection (keep_alive = 180 s)
ggml_metal_device_init: GPU name: Apple M1
ggml_metal_device_init: GPU family: MTLGPUFamilyApple7 (1007)
ggml_metal_device_init: GPU family: MTLGPUFamilyCommon3 (3003)
ggml_metal_device_init: GPU family: MTLGPUFamilyMetal3 (5001)
ggml_metal_device_init: simdgroup reduction = true
ggml_metal_device_init: simdgroup matrix mul. = true
ggml_metal_device_init: has unified memory = true
ggml_metal_device_init: has bfloat = true
ggml_metal_device_init: has tensor = false
ggml_metal_device_init: use residency sets = true
ggml_metal_device_init: use shared buffers = true
ggml_metal_device_init: recommendedMaxWorkingSetSize = 11453.25 MB
whisper_init_with_params_no_state: devices = 3
whisper_init_with_params_no_state: backends = 3
whisper_model_load: loading model
whisper_model_load: n_vocab = 51865
whisper_model_load: n_audio_ctx = 1500
whisper_model_load: n_audio_state = 384
whisper_model_load: n_audio_head = 6
whisper_model_load: n_audio_layer = 4
whisper_model_load: n_text_ctx = 448
whisper_model_load: n_text_state = 384
whisper_model_load: n_text_head = 6
whisper_model_load: n_text_layer = 4
whisper_model_load: n_mels = 80
whisper_model_load: ftype = 1
whisper_model_load: qntvr = 0
whisper_model_load: type = 1 (tiny)
whisper_model_load: adding 1608 extra tokens
whisper_model_load: n_langs = 99
whisper_model_load: Metal total size = 77.11 MB
whisper_model_load: model size = 77.11 MB
2026-04-09T19:54:07.178366Z INFO screenpipe_audio::transcription::engine: whisper model loaded successfully
2026-04-09T19:54:07.178441Z INFO screenpipe_audio::audio_manager::device_monitor: model capabilities changed after background download; restarting central handlers
2026-04-09T19:54:07.178636Z INFO screenpipe_audio::transcription::engine: whisper model available: "/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin"
2026-04-09T19:54:07.178655Z INFO screenpipe_audio::transcription::whisper::model: whisper context: gpu acceleration enabled (Metal on macOS, Vulkan on Windows)
2026-04-09T19:54:07.178661Z INFO screenpipe_audio::transcription::engine: loading whisper model with GPU acceleration...
whisper_init_from_file_with_params_no_state: loading model from '/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin'
whisper_init_with_params_no_state: use gpu = 1
whisper_init_with_params_no_state: flash attn = 0
whisper_init_with_params_no_state: gpu_device = 0
whisper_init_with_params_no_state: dtw = 0
whisper_init_with_params_no_state: devices = 3
whisper_init_with_params_no_state: backends = 3
whisper_model_load: loading model
whisper_model_load: n_vocab = 51865
whisper_model_load: n_audio_ctx = 1500
whisper_model_load: n_audio_state = 384
whisper_model_load: n_audio_head = 6
whisper_model_load: n_audio_layer = 4
whisper_model_load: n_text_ctx = 448
whisper_model_load: n_text_state = 384
whisper_model_load: n_text_head = 6
whisper_model_load: n_text_layer = 4
whisper_model_load: n_mels = 80
whisper_model_load: ftype = 1
whisper_model_load: qntvr = 0
whisper_model_load: type = 1 (tiny)
whisper_model_load: adding 1608 extra tokens
whisper_model_load: n_langs = 99
whisper_model_load: Metal total size = 77.11 MB
whisper_model_load: model size = 77.11 MB
2026-04-09T19:54:07.242116Z INFO screenpipe_audio::transcription::engine: whisper model loaded successfully
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
2026-04-09T19:54:07.250010Z INFO screenpipe_audio::audio_manager::manager: transcription session created (will be reused across segments)
2026-04-09T19:54:07.250055Z INFO screenpipe_audio::audio_manager::manager: central audio-receiver handler restarted for capability refresh
2026-04-09T19:54:07.250066Z INFO screenpipe_audio::audio_manager::manager: central transcription-receiver handler restarted for capability refresh
2026-04-09T19:54:41.976071Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-1231441521457939656, trigger=visual_change)
DOCKER
Close Tab
DEV (-zsh)
Close Tab
APP (-zsh)
Close Tab
screenpipe"
Close Tab
Python"
Close Tab
⌥⌘1
screenpipe"...
|
iTerm2
|
screenpipe"
|
NULL
|
|
Last login: Thu Apr 9 19:52:08 on ttys007
Poetry Last login: Thu Apr 9 19:52:08 on ttys007
Poetry could not find a pyproject.toml file in /Users/lukas/jiminny/app or its parents
Poetry could not find a pyproject.toml file in /Users/lukas/jiminny/app or its parents
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/jiminny/app (JY-18909-automated-reports-ask-jiminny) $ cd
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record
detected hardware tier: Mid
checking permissions...
screen recording: ok
microphone: ok
accessibility: ok
2026-04-09T19:53:07.212588Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T19:53:08.176294Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T19:53:08.180179Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T19:53:08.180172Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T19:53:08.180644Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T19:53:08.181079Z INFO screenpipe_audio::speaker::models: initiating wespeaker_en_voxceleb_CAM++.onnx model download...
2026-04-09T19:53:08.181124Z WARN screenpipe_audio::segmentation::segmentation_manager: embedding model unavailable at startup: wespeaker_en_voxceleb_CAM++.onnx model not available yet; download started in background
2026-04-09T19:53:08.181175Z INFO screenpipe_audio::speaker::models: initiating segmentation-3.0.onnx model download...
2026-04-09T19:53:08.181183Z WARN screenpipe_audio::segmentation::segmentation_manager: segmentation model unavailable at startup: segmentation-3.0.onnx model not available yet; download started in background
2026-04-09T19:53:08.181157Z INFO screenpipe_audio::speaker::models: wespeaker_en_voxceleb_CAM++.onnx model download attempt 1/3
2026-04-09T19:53:08.181223Z WARN screenpipe_audio::audio_manager::manager: silero vad unavailable, falling back to webrtc: silero vad model not available yet; download started in background
2026-04-09T19:53:08.181225Z INFO screenpipe_audio::speaker::models: downloading wespeaker_en_voxceleb_CAM++.onnx model from https://github.com/screenpipe/screenpipe/raw/refs/heads/main/crates/screenpipe-audio/models/pyannote/wespeaker_en_voxceleb_CAM++.onnx
2026-04-09T19:53:08.181212Z INFO screenpipe_audio::speaker::models: segmentation-3.0.onnx model download attempt 1/3
2026-04-09T19:53:08.181526Z INFO screenpipe_audio::speaker::models: downloading segmentation-3.0.onnx model from https://github.com/screenpipe/screenpipe/raw/refs/heads/main/crates/screenpipe-audio/models/pyannote/segmentation-3.0.onnx
2026-04-09T19:53:08.181252Z INFO screenpipe_audio::vad::silero: downloading Silero VAD v5 model...
2026-04-09T19:53:08.181607Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T19:53:08.181646Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T19:53:08.181366Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T19:53:08.181759Z INFO screenpipe_core::pipes: installed built-in pipe: day-recap
2026-04-09T19:53:08.181952Z INFO screenpipe_core::pipes: installed built-in pipe: standup-update
2026-04-09T19:53:08.182315Z INFO screenpipe_core::pipes: installed built-in pipe: ai-habits
2026-04-09T19:53:08.182501Z INFO screenpipe_core::pipes: installed built-in pipe: time-breakdown
2026-04-09T19:53:08.182680Z INFO screenpipe_core::pipes: installed built-in pipe: video-export
2026-04-09T19:53:08.182875Z INFO screenpipe_core::pipes: installed built-in pipe: meeting-summary
2026-04-09T19:53:08.183091Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T19:53:08.183146Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T19:53:08.183187Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T19:53:08.183237Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T19:53:08.183306Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T19:53:08.183363Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T19:53:08.183380Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ false │
│ vision disabled │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ soundcore AeroClip (input) │
│ │ System Audio (output) │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: https://github.com/screenpipe/screenpipe/releases
2026-04-09T19:53:08.185865Z INFO screenpipe: starting UI event capture
2026-04-09T19:53:08.185586Z INFO screenpipe_core::pipes: pipe scheduler started (generation 2)
2026-04-09T19:53:08.188922Z WARN screenpipe: pi agent install failed: bun not found — install from https://bun.sh
2026-04-09T19:53:08.194461Z INFO screenpipe_engine::power::manager: initial power profile: Performance (on_ac=true, battery=Some(100))
2026-04-09T19:53:08.201212Z INFO screenpipe_engine::ui_recorder: Starting UI event capture
2026-04-09T19:53:08.217633Z INFO screenpipe_engine::calendar_speaker_id: speaker identification: started (user_name=<not set>)
2026-04-09T19:53:08.217794Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warming from DB (2026-04-08 16:53:08.217792 UTC to 2026-04-09 16:53:08.217792 UTC)
2026-04-09T19:53:08.217649Z INFO screenpipe_engine::ui_recorder: UI recording session started: 3b59b1d1-d80e-4689-8c72-6476af4571df
2026-04-09T19:53:08.220368Z INFO screenpipe_engine::meeting_detector: meeting v2: detection loop started (base_interval=5s, profiles=12)
2026-04-09T19:53:08.222489Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warmed with 0 frame entries, coverage from 2026-04-08 16:53:08.217792 UTC
2026-04-09T19:53:08.227257Z INFO screenpipe_engine::server: Server listening on [IP_ADDRESS]:3030
2026-04-09T19:53:08.232250Z INFO screenpipe_connect::mdns: mdns: advertising screenpipe on port 3030
2026-04-09T19:53:08.543211Z INFO screenpipe_engine::vision_manager::manager: Starting vision recording for monitor 1 (1440x900)
2026-04-09T19:53:08.543241Z INFO screenpipe_engine::vision_manager::manager: Starting event-driven capture for monitor 1 (device: monitor_1)
2026-04-09T19:53:08.543253Z INFO screenpipe_engine::vision_manager::manager: Skipping monitor 2 (Display 2_3008x1253_-813,-1253) — not in allowed list
2026-04-09T19:53:08.543262Z INFO screenpipe_engine::vision_manager::monitor_watcher: Starting monitor watcher (polling every 5 seconds)
2026-04-09T19:53:08.543299Z INFO screenpipe_engine::event_driven_capture: event-driven capture started for monitor 1 (device: monitor_1)
2026-04-09T19:53:09.210453Z INFO screenpipe_audio::vad::silero: Silero VAD v5 model downloaded and saved to: "/Users/lukas/Library/Caches/screenpipe/vad/silero_vad_v5.onnx"
2026-04-09T19:53:09.352790Z INFO screenpipe_audio::speaker::models: saving segmentation-3.0.onnx model (5983836 bytes) to "/Users/lukas/Library/Caches/screenpipe/models/segmentation-3.0.onnx"
2026-04-09T19:53:09.355799Z INFO screenpipe_audio::speaker::models: segmentation-3.0.onnx model successfully downloaded and saved
2026-04-09T19:53:09.367746Z INFO sck_rs::stream_manager: persistent SCK stream started for display 1 (1440x900, 2fps)
2026-04-09T19:53:10.102366Z INFO screenpipe_audio::speaker::models: saving wespeaker_en_voxceleb_CAM++.onnx model (29292684 bytes) to "/Users/lukas/Library/Caches/screenpipe/models/wespeaker_en_voxceleb_CAM++.onnx"
2026-04-09T19:53:10.112264Z INFO screenpipe_audio::speaker::models: wespeaker_en_voxceleb_CAM++.onnx model successfully downloaded and saved
2026-04-09T19:53:13.221908Z INFO screenpipe_engine::retention: retention: initialized with 14d retention
2026-04-09T19:53:13.222855Z INFO screenpipe: local retention auto-enabled (14 days)
2026-04-09T19:53:15.841867Z INFO screenpipe_engine::event_driven_capture: startup capture for monitor 1: frame_id=1, dur=797ms
2026-04-09T19:53:18.189478Z WARN screenpipe_audio::transcription::engine: whisper model is not available locally yet for WhisperTiny; audio transcription disabled until download completes
2026-04-09T19:53:18.189794Z INFO screenpipe_audio::audio_manager::manager: transcription session created (will be reused across segments)
2026-04-09T19:53:18.190227Z INFO screenpipe_audio::transcription::whisper::model: downloading model "ggml-tiny.bin"
2026-04-09T19:53:18.190847Z INFO screenpipe_audio::audio_manager::manager: audio manager started
2026-04-09T19:53:18.190916Z INFO screenpipe_audio::audio_manager::manager: calendar-assisted speaker diarization: listening for meeting events
ggml-tiny.bin [00:00:01] [███████████████████████████████████████████████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] 38.34 MiB/74.09 MiB 21.65 MiB/s (2s)2026-04-09T19:53:19.720503Z INFO screenpipe_audio::audio_manager::device_monitor: model capabilities changed after background download; restarting central handlers
2026-04-09T19:53:19.720675Z WARN screenpipe_audio::transcription::engine: whisper model is not available locally yet for WhisperTiny; audio transcription disabled until download completes
2026-04-09T19:53:19.720732Z INFO screenpipe_audio::audio_manager::manager: transcription session created (will be reused across segments)
2026-04-09T19:53:19.720787Z INFO screenpipe_audio::audio_manager::manager: central audio-receiver handler restarted for capability refresh
2026-04-09T19:53:19.720795Z INFO screenpipe_audio::audio_manager::manager: central transcription-receiver handler restarted for capability refresh
2026-04-09T19:53:19.721204Z INFO screenpipe_audio::transcription::whisper::model: downloading model "ggml-tiny.bin"
ggml-tiny.bin [00:00:01] [██████████████████████████████████████████████████████████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] 45.66 MiB/74.09 MiB 23.13 MiB/s (1s)
ggml-tiny.bin [00:00:00] [███░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] 2.09 MiB/74.09 MiB 1.63 MiB/s (44s)2026-04-09T19:53:20.383627Z INFO screenpipe_audio::device::device_manager: starting recording for device: System Audio (output)
2026-04-09T19:53:20.383837Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
ggml-tiny.bin [00:00:01] [██████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] 9.22 MiB/74.09 MiB 4.95 MiB/s (13s)2026-04-09T19:53:20.972014Z INFO screenpipe_audio::device::device_manager: starting recording for device: soundcore AeroClip (input)
2026-04-09T19:53:20.972101Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for soundcore AeroClip (input) (30s segments)
ggml-tiny.bin [00:00:02] [███████████████████████████████████████████████████████████████████████████████████████████████████████████████████] 74.09 MiB/74.09 MiB 28.28 MiB/s (0s)2026-04-09T19:53:21.047457Z INFO screenpipe_audio::transcription::whisper::model: model downloaded ggml-tiny.bin
2026-04-09T19:53:21.047589Z INFO screenpipe_audio::transcription::engine: whisper model downloaded in background: "/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin"
ggml-tiny.bin [00:00:03] [███████████████████████████████████████████████████████████████████████████████████████████████████████████████████] 74.09 MiB/74.09 MiB 23.56 MiB/s (0s)2026-04-09T19:53:23.080110Z INFO screenpipe_audio::transcription::whisper::model: model downloaded ggml-tiny.bin
2026-04-09T19:53:23.080240Z INFO screenpipe_audio::transcription::engine: whisper model downloaded in background: "/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin"
2026-04-09T19:53:52.431282Z INFO screenpipe_audio::transcription::engine: whisper model available: "/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin"
2026-04-09T19:53:52.431325Z INFO screenpipe_audio::transcription::whisper::model: whisper context: gpu acceleration enabled (Metal on macOS, Vulkan on Windows)
2026-04-09T19:53:52.431330Z INFO screenpipe_audio::transcription::engine: loading whisper model with GPU acceleration...
whisper_init_from_file_with_params_no_state: loading model from '/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin'
whisper_init_with_params_no_state: use gpu = 1
whisper_init_with_params_no_state: flash attn = 0
whisper_init_with_params_no_state: gpu_device = 0
whisper_init_with_params_no_state: dtw = 0
ggml_metal_device_init: tensor API disabled for pre-M5 and pre-A19 devices
ggml_metal_library_init: using embedded metal library
ggml_metal_library_init: loaded in 14.652 sec
ggml_metal_rsets_init: creating a residency set collection (keep_alive = 180 s)
ggml_metal_device_init: GPU name: Apple M1
ggml_metal_device_init: GPU family: MTLGPUFamilyApple7 (1007)
ggml_metal_device_init: GPU family: MTLGPUFamilyCommon3 (3003)
ggml_metal_device_init: GPU family: MTLGPUFamilyMetal3 (5001)
ggml_metal_device_init: simdgroup reduction = true
ggml_metal_device_init: simdgroup matrix mul. = true
ggml_metal_device_init: has unified memory = true
ggml_metal_device_init: has bfloat = true
ggml_metal_device_init: has tensor = false
ggml_metal_device_init: use residency sets = true
ggml_metal_device_init: use shared buffers = true
ggml_metal_device_init: recommendedMaxWorkingSetSize = 11453.25 MB
whisper_init_with_params_no_state: devices = 3
whisper_init_with_params_no_state: backends = 3
whisper_model_load: loading model
whisper_model_load: n_vocab = 51865
whisper_model_load: n_audio_ctx = 1500
whisper_model_load: n_audio_state = 384
whisper_model_load: n_audio_head = 6
whisper_model_load: n_audio_layer = 4
whisper_model_load: n_text_ctx = 448
whisper_model_load: n_text_state = 384
whisper_model_load: n_text_head = 6
whisper_model_load: n_text_layer = 4
whisper_model_load: n_mels = 80
whisper_model_load: ftype = 1
whisper_model_load: qntvr = 0
whisper_model_load: type = 1 (tiny)
whisper_model_load: adding 1608 extra tokens
whisper_model_load: n_langs = 99
whisper_model_load: Metal total size = 77.11 MB
whisper_model_load: model size = 77.11 MB
2026-04-09T19:54:07.178366Z INFO screenpipe_audio::transcription::engine: whisper model loaded successfully
2026-04-09T19:54:07.178441Z INFO screenpipe_audio::audio_manager::device_monitor: model capabilities changed after background download; restarting central handlers
2026-04-09T19:54:07.178636Z INFO screenpipe_audio::transcription::engine: whisper model available: "/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin"
2026-04-09T19:54:07.178655Z INFO screenpipe_audio::transcription::whisper::model: whisper context: gpu acceleration enabled (Metal on macOS, Vulkan on Windows)
2026-04-09T19:54:07.178661Z INFO screenpipe_audio::transcription::engine: loading whisper model with GPU acceleration...
whisper_init_from_file_with_params_no_state: loading model from '/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin'
whisper_init_with_params_no_state: use gpu = 1
whisper_init_with_params_no_state: flash attn = 0
whisper_init_with_params_no_state: gpu_device = 0
whisper_init_with_params_no_state: dtw = 0
whisper_init_with_params_no_state: devices = 3
whisper_init_with_params_no_state: backends = 3
whisper_model_load: loading model
whisper_model_load: n_vocab = 51865
whisper_model_load: n_audio_ctx = 1500
whisper_model_load: n_audio_state = 384
whisper_model_load: n_audio_head = 6
whisper_model_load: n_audio_layer = 4
whisper_model_load: n_text_ctx = 448
whisper_model_load: n_text_state = 384
whisper_model_load: n_text_head = 6
whisper_model_load: n_text_layer = 4
whisper_model_load: n_mels = 80
whisper_model_load: ftype = 1
whisper_model_load: qntvr = 0
whisper_model_load: type = 1 (tiny)
whisper_model_load: adding 1608 extra tokens
whisper_model_load: n_langs = 99
whisper_model_load: Metal total size = 77.11 MB
whisper_model_load: model size = 77.11 MB
2026-04-09T19:54:07.242116Z INFO screenpipe_audio::transcription::engine: whisper model loaded successfully
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
2026-04-09T19:54:07.250010Z INFO screenpipe_audio::audio_manager::manager: transcription session created (will be reused across segments)
2026-04-09T19:54:07.250055Z INFO screenpipe_audio::audio_manager::manager: central audio-receiver handler restarted for capability refresh
2026-04-09T19:54:07.250066Z INFO screenpipe_audio::audio_manager::manager: central transcription-receiver handler restarted for capability refresh
2026-04-09T19:54:41.976071Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-1231441521457939656, trigger=visual_change)
DOCKER
Close Tab
DEV (-zsh)
Close Tab
APP (-zsh)
Close Tab
screenpipe"
Close Tab
-zsh
Close Tab
⌥⌘1
screenpipe"...
|
iTerm2
|
screenpipe"
|
NULL
|
|
ggml_metal_free: deallocating
whisper_backend_init ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:09:30.413642Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:11:31.943416Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
2026-04-09T20:12:23.632839Z WARN screenpipe_audio::core::run_record_and_transcribe: no audio received from System Audio (output) for 30s - stream dead, triggering reconnect
2026-04-09T20:12:24.634524Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:24.634717Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:25.636167Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:25.636241Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:26.637475Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:26.637632Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:27.639481Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:27.639674Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:28.641392Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:28.641468Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:29.643401Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:29.643465Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:30.645555Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:30.645736Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:31.646650Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:31.646804Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:32.648705Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:32.648787Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:33.649937Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:33.650126Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:34.652420Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:34.652503Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:35.654363Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:35.654429Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:36.655671Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:36.655835Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:37.657385Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:37.657510Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:38.659335Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:38.659397Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:39.660525Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:39.660601Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:40.662552Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:40.662717Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:41.664366Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:41.664429Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:42.666360Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:42.666464Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:43.668128Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:43.668333Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:44.670229Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:44.670277Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:45.671405Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:45.671453Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:46.672748Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:46.672859Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:47.673601Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:47.673766Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:48.676092Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:48.676148Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:49.677349Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)...
|
iTerm2
|
screenpipe"
|
NULL
|
|
whisper_backend_init_gpu: found GPU device 0: Meta whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:09:30.413642Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:11:31.943416Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
2026-04-09T20:12:23.632839Z WARN screenpipe_audio::core::run_record_and_transcribe: no audio received from System Audio (output) for 30s - stream dead, triggering reconnect
2026-04-09T20:12:24.634524Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:24.634717Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:25.636167Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:25.636241Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:26.637475Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:26.637632Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:27.639481Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:27.639674Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:28.641392Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:28.641468Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:29.643401Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:29.643465Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:30.645555Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:30.645736Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:31.646650Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:31.646804Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:32.648705Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:32.648787Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:33.649937Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:33.650126Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:34.652420Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:34.652503Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:35.654363Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:35.654429Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:36.655671Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:36.655835Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:37.657385Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:37.657510Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:38.659335Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:38.659397Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:39.660525Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:39.660601Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:40.662552Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:40.662717Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:41.664366Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:41.664429Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:42.666360Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:42.666464Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:43.668128Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:43.668333Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:44.670229Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:44.670277Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:45.671405Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:45.671453Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:46.672748Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:46.672859Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:47.673601Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:47.673766Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:48.676092Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:48.676148Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:49.677349Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:49.677498Z INFO screenpipe_audio::core::run_record_and_transcri...
|
iTerm2
|
screenpipe"
|
NULL
|
|
whisper_init_state: compute buffer (encode) = 65 whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:11:31.943416Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
2026-04-09T20:12:23.632839Z WARN screenpipe_audio::core::run_record_and_transcribe: no audio received from System Audio (output) for 30s - stream dead, triggering reconnect
2026-04-09T20:12:24.634524Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:24.634717Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:25.636167Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:25.636241Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:26.637475Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:26.637632Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:27.639481Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:27.639674Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:28.641392Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:28.641468Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:29.643401Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:29.643465Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:30.645555Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:30.645736Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:31.646650Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:31.646804Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:32.648705Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:32.648787Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:33.649937Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:33.650126Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:34.652420Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:34.652503Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:35.654363Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:35.654429Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:36.655671Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:36.655835Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:37.657385Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:37.657510Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:38.659335Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:38.659397Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:39.660525Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:39.660601Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:40.662552Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:40.662717Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:41.664366Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:41.664429Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:42.666360Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:42.666464Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:43.668128Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:43.668333Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:44.670229Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:44.670277Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:45.671405Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:45.671453Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:46.672748Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:46.672859Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:47.673601Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:47.673766Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:48.676092Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:48.676148Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:49.677349Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:49.677498Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:50.679547Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:50.679705Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:51.681540Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:51.681638Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:52.682700Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:52.682893Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:53.683432Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:53.683587Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:54.685305Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:54.685380Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:55.687119Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:55.687360Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:56.688435Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:56.688513Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:57.690272Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:57.690340Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:58.692254Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:58.692310Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:59.693320Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:59.693493Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:00.694348Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:00.694433Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:01.696392Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:01.696444Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:02.698281Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:02.698458Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:03.700030Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:03.700146Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:04.702226Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:04.702294Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:05.704281Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:05.704359Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:06.705808Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:06.705920Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:07.707231Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:07.707298Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:08.182408Z INFO screenpipe_engine::server: api_usage_5min: 2 requests
2026-04-09T20:13:08.708393Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:08.708510Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026...
|
iTerm2
|
screenpipe"
|
NULL
|
|
whisper_init_state: compute buffer (cross) = 8 whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:11:31.943416Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
2026-04-09T20:12:23.632839Z WARN screenpipe_audio::core::run_record_and_transcribe: no audio received from System Audio (output) for 30s - stream dead, triggering reconnect
2026-04-09T20:12:24.634524Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:24.634717Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:25.636167Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:25.636241Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:26.637475Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:26.637632Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:27.639481Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:27.639674Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:28.641392Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:28.641468Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:29.643401Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:29.643465Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:30.645555Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:30.645736Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:31.646650Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:31.646804Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:32.648705Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:32.648787Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:33.649937Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:33.650126Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:34.652420Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:34.652503Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:35.654363Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:35.654429Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:36.655671Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:36.655835Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:37.657385Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:37.657510Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:38.659335Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:38.659397Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:39.660525Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:39.660601Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:40.662552Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:40.662717Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:41.664366Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:41.664429Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:42.666360Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:42.666464Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:43.668128Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:43.668333Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:44.670229Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:44.670277Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:45.671405Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:45.671453Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:46.672748Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:46.672859Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:47.673601Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:47.673766Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:48.676092Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:48.676148Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:49.677349Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:49.677498Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:50.679547Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:50.679705Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:51.681540Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:51.681638Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:52.682700Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:52.682893Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:53.683432Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:53.683587Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:54.685305Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:54.685380Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:55.687119Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:55.687360Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:56.688435Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:56.688513Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:57.690272Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:57.690340Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:58.692254Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:58.692310Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:59.693320Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:59.693493Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:00.694348Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:00.694433Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:01.696392Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:01.696444Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:02.698281Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:02.698458Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:03.700030Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:03.700146Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:04.702226Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:04.702294Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:05.704281Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:05.704359Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:06.705808Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:06.705920Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:07.707231Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:07.707298Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:08.182408Z INFO screenpipe_engine::server: api_usage_5min: 2 requests
2026-04-09T20:13:08.708393Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:08.708510Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:09.710915Z INFO screenpipe_audio::core::run...
|
iTerm2
|
screenpipe"
|
NULL
|
|
whisper_init_state: compute buffer (decode) = 96 whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:11:31.943416Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
2026-04-09T20:12:23.632839Z WARN screenpipe_audio::core::run_record_and_transcribe: no audio received from System Audio (output) for 30s - stream dead, triggering reconnect
2026-04-09T20:12:24.634524Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:24.634717Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:25.636167Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:25.636241Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:26.637475Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:26.637632Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:27.639481Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:27.639674Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:28.641392Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:28.641468Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:29.643401Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:29.643465Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:30.645555Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:30.645736Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:31.646650Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:31.646804Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:32.648705Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:32.648787Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:33.649937Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:33.650126Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:34.652420Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:34.652503Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:35.654363Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:35.654429Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:36.655671Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:36.655835Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:37.657385Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:37.657510Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:38.659335Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:38.659397Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:39.660525Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:39.660601Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:40.662552Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:40.662717Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:41.664366Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:41.664429Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:42.666360Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:42.666464Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:43.668128Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:43.668333Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:44.670229Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:44.670277Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:45.671405Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:45.671453Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:46.672748Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:46.672859Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:47.673601Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:47.673766Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:48.676092Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:48.676148Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:49.677349Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:49.677498Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:50.679547Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:50.679705Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:51.681540Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:51.681638Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:52.682700Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:52.682893Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:53.683432Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:53.683587Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:54.685305Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:54.685380Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:55.687119Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:55.687360Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:56.688435Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:56.688513Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:57.690272Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:57.690340Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:58.692254Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:58.692310Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:59.693320Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:59.693493Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:00.694348Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:00.694433Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:01.696392Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:01.696444Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:02.698281Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:02.698458Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:03.700030Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:03.700146Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:04.702226Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:04.702294Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:05.704281Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:05.704359Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:06.705808Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:06.705920Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:07.707231Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:07.707298Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:08.182408Z INFO screenpipe_engine::server: api_usage_5min: 2 requests
2026-04-09T20:13:08.708393Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:08.708510Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:09.710915Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:09.710974Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:10.712639Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:10.712711Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:11.714135Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:11.714182Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:12.716251Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:12.716313Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:13.717538Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:13.717682Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:14.719220Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:14.719306Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:15.721206Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:15.721607Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:16.723013Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:16.723167Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:17.725383Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:17.725436Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output...
|
iTerm2
|
screenpipe"
|
NULL
|
|
whisper_backend_init_gpu: using Metal backend
ggml whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:11:31.943416Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
2026-04-09T20:12:23.632839Z WARN screenpipe_audio::core::run_record_and_transcribe: no audio received from System Audio (output) for 30s - stream dead, triggering reconnect
2026-04-09T20:12:24.634524Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:24.634717Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:25.636167Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:25.636241Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:26.637475Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:26.637632Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:27.639481Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:27.639674Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:28.641392Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:28.641468Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:29.643401Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:29.643465Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:30.645555Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:30.645736Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:31.646650Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:31.646804Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:32.648705Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:32.648787Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:33.649937Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:33.650126Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:34.652420Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:34.652503Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:35.654363Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:35.654429Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:36.655671Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:36.655835Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:37.657385Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:37.657510Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:38.659335Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:38.659397Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:39.660525Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:39.660601Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:40.662552Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:40.662717Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:41.664366Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:41.664429Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:42.666360Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:42.666464Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:43.668128Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:43.668333Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:44.670229Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:44.670277Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:45.671405Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:45.671453Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:46.672748Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:46.672859Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:47.673601Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:47.673766Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:48.676092Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:48.676148Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:49.677349Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:49.677498Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:50.679547Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:50.679705Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:51.681540Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:51.681638Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:52.682700Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:52.682893Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:53.683432Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:53.683587Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:54.685305Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:54.685380Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:55.687119Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:55.687360Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:56.688435Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:56.688513Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:57.690272Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:57.690340Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:58.692254Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:58.692310Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:59.693320Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:59.693493Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:00.694348Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:00.694433Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:01.696392Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:01.696444Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:02.698281Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:02.698458Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:03.700030Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:03.700146Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:04.702226Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:04.702294Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:05.704281Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:05.704359Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:06.705808Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:06.705920Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:07.707231Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:07.707298Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:08.182408Z INFO screenpipe_engine::server: api_usage_5min: 2 requests
2026-04-09T20:13:08.708393Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:08.708510Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:09.710915Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:09.710974Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:10.712639Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:10.712711Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:11.714135Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:11.714182Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:12.716251Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:12.716313Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:13.717538Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:13.717682Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:14.719220Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:14.719306Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:15.721206Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:15.721607Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:16.723013Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:16.723167Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:17.725383Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:17.725436Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:18.727254Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:18.727423Z INFO sc...
|
iTerm2
|
screenpipe"
|
NULL
|
|
ggml_metal_init: allocating
ggml_metal_init: found ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:11:31.943416Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
2026-04-09T20:12:23.632839Z WARN screenpipe_audio::core::run_record_and_transcribe: no audio received from System Audio (output) for 30s - stream dead, triggering reconnect
2026-04-09T20:12:24.634524Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:24.634717Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:25.636167Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:25.636241Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:26.637475Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:26.637632Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:27.639481Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:27.639674Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:28.641392Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:28.641468Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:29.643401Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:29.643465Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:30.645555Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:30.645736Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:31.646650Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:31.646804Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:32.648705Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:32.648787Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:33.649937Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:33.650126Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:34.652420Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:34.652503Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:35.654363Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:35.654429Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:36.655671Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:36.655835Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:37.657385Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:37.657510Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:38.659335Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:38.659397Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:39.660525Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:39.660601Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:40.662552Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:40.662717Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:41.664366Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:41.664429Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:42.666360Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:42.666464Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:43.668128Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:43.668333Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:44.670229Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:44.670277Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:45.671405Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:45.671453Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:46.672748Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:46.672859Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:47.673601Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:47.673766Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:48.676092Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:48.676148Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:49.677349Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:49.677498Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:50.679547Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:50.679705Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:51.681540Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:51.681638Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:52.682700Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:52.682893Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:53.683432Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:53.683587Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:54.685305Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:54.685380Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:55.687119Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:55.687360Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:56.688435Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:56.688513Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:57.690272Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:57.690340Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:58.692254Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:58.692310Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:59.693320Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:59.693493Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:00.694348Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:00.694433Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:01.696392Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:01.696444Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:02.698281Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:02.698458Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:03.700030Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:03.700146Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:04.702226Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:04.702294Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:05.704281Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:05.704359Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:06.705808Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:06.705920Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:07.707231Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:07.707298Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:08.182408Z INFO screenpipe_engine::server: api_usage_5min: 2 requests
2026-04-09T20:13:08.708393Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:08.708510Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:09.710915Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:09.710974Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:10.712639Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:10.712711Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:11.714135Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:11.714182Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:12.716251Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:12.716313Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:13.717538Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:13.717682Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:14.719220Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:14.719306Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:15.721206Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:15.721607Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:16.723013Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:16.723167Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:17.725383Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:17.725436Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:18.727254Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:18.727423Z INFO screenpipe_audio::core::run_record_and_transcrib...
|
iTerm2
|
screenpipe"
|
NULL
|
|
ggml_metal_init: found device: Apple M1
ggml_metal ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:11:31.943416Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
2026-04-09T20:12:23.632839Z WARN screenpipe_audio::core::run_record_and_transcribe: no audio received from System Audio (output) for 30s - stream dead, triggering reconnect
2026-04-09T20:12:24.634524Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:24.634717Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:25.636167Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:25.636241Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:26.637475Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:26.637632Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:27.639481Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:27.639674Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:28.641392Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:28.641468Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:29.643401Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:29.643465Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:30.645555Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:30.645736Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:31.646650Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:31.646804Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:32.648705Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:32.648787Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:33.649937Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:33.650126Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:34.652420Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:34.652503Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:35.654363Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:35.654429Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:36.655671Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:36.655835Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:37.657385Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:37.657510Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:38.659335Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:38.659397Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:39.660525Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:39.660601Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:40.662552Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:40.662717Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:41.664366Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:41.664429Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:42.666360Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:42.666464Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:43.668128Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:43.668333Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:44.670229Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:44.670277Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:45.671405Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:45.671453Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:46.672748Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:46.672859Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:47.673601Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:47.673766Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:48.676092Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:48.676148Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:49.677349Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:49.677498Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:50.679547Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:50.679705Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:51.681540Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:51.681638Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:52.682700Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:52.682893Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:53.683432Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:53.683587Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:54.685305Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:54.685380Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:55.687119Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:55.687360Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:56.688435Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:56.688513Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:57.690272Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:57.690340Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:58.692254Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:58.692310Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:59.693320Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:59.693493Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:00.694348Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:00.694433Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:01.696392Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:01.696444Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:02.698281Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:02.698458Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:03.700030Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:03.700146Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:04.702226Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:04.702294Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:05.704281Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:05.704359Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:06.705808Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:06.705920Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:07.707231Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:07.707298Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:08.182408Z INFO screenpipe_engine::server: api_usage_5min: 2 requests
2026-04-09T20:13:08.708393Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:08.708510Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:09.710915Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:09.710974Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:10.712639Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:10.712711Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:11.714135Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:11.714182Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:12.716251Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:12.716313Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:13.717538Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:13.717682Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:14.719220Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:14.719306Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:15.721206Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:15.721607Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:16.723013Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:16.723167Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:17.725383Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:17.725436Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:18.727254Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:18.727423Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for Sys...
|
iTerm2
|
screenpipe"
|
NULL
|
|
ggml_metal_init: use concurrency = true
ggml_me ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:11:31.943416Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
2026-04-09T20:12:23.632839Z WARN screenpipe_audio::core::run_record_and_transcribe: no audio received from System Audio (output) for 30s - stream dead, triggering reconnect
2026-04-09T20:12:24.634524Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:24.634717Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:25.636167Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:25.636241Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:26.637475Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:26.637632Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:27.639481Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:27.639674Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:28.641392Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:28.641468Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:29.643401Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:29.643465Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:30.645555Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:30.645736Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:31.646650Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:31.646804Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:32.648705Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:32.648787Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:33.649937Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:33.650126Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:34.652420Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:34.652503Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:35.654363Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:35.654429Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:36.655671Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:36.655835Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:37.657385Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:37.657510Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:38.659335Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:38.659397Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:39.660525Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:39.660601Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:40.662552Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:40.662717Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:41.664366Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:41.664429Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:42.666360Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:42.666464Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:43.668128Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:43.668333Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:44.670229Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:44.670277Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:45.671405Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:45.671453Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:46.672748Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:46.672859Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:47.673601Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:47.673766Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:48.676092Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:48.676148Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:49.677349Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:49.677498Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:50.679547Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:50.679705Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:51.681540Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:51.681638Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:52.682700Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:52.682893Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:53.683432Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:53.683587Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:54.685305Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:54.685380Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:55.687119Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:55.687360Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:56.688435Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:56.688513Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:57.690272Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:57.690340Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:58.692254Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:58.692310Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:12:59.693320Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:12:59.693493Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:00.694348Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:00.694433Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:01.696392Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:01.696444Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:02.698281Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:02.698458Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:03.700030Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:03.700146Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:04.702226Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:04.702294Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:05.704281Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:05.704359Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:06.705808Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:06.705920Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:07.707231Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:07.707298Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:08.182408Z INFO screenpipe_engine::server: api_usage_5min: 2 requests
2026-04-09T20:13:08.708393Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:08.708510Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:09.710915Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:09.710974Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:10.712639Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:10.712711Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:11.714135Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:11.714182Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:12.716251Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:12.716313Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:13.717538Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:13.717682Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:14.719220Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:14.719306Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:15.721206Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:15.721607Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:16.723013Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:16.723167Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:17.725383Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:17.725436Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:18.727254Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:13:18.727423Z INFO screenpipe_audio::core::run_record_and_transcribe: stopped recording for System Audio (output) (disconnected)
2026-04-09T20:13:19.728646Z INFO screenpipe_audio::core::run_record_and_transcribe: starting conti...
|
iTerm2
|
screenpipe"
|
NULL
|
|
ggml_metal_device_init: simdgroup reduction = tr ggml_metal_device_init: simdgroup reduction = true
ggml_metal_device_init: simdgroup matrix mul. = true
ggml_metal_device_init: has unified memory = true
ggml_metal_device_init: has bfloat = true
ggml_metal_device_init: has tensor = false
ggml_metal_device_init: use residency sets = true
ggml_metal_device_init: use shared buffers = true
ggml_metal_device_init: recommendedMaxWorkingSetSize = 11453.25 MB
whisper_init_with_params_no_state: devices = 3
whisper_init_with_params_no_state: backends = 3
whisper_model_load: loading model
whisper_model_load: n_vocab = 51865
whisper_model_load: n_audio_ctx = 1500
whisper_model_load: n_audio_state = 384
whisper_model_load: n_audio_head = 6
whisper_model_load: n_audio_layer = 4
whisper_model_load: n_text_ctx = 448
whisper_model_load: n_text_state = 384
whisper_model_load: n_text_head = 6
whisper_model_load: n_text_layer = 4
whisper_model_load: n_mels = 80
whisper_model_load: ftype = 1
whisper_model_load: qntvr = 0
whisper_model_load: type = 1 (tiny)
whisper_model_load: adding 1608 extra tokens
whisper_model_load: n_langs = 99
whisper_model_load: Metal total size = 77.11 MB
whisper_model_load: model size = 77.11 MB
2026-04-09T20:14:54.059372Z INFO screenpipe_audio::transcription::engine: whisper model loaded successfully
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
2026-04-09T20:14:54.062532Z INFO screenpipe_audio::audio_manager::manager: transcription session created (will be reused across segments)
2026-04-09T20:14:54.062898Z INFO screenpipe_audio::audio_manager::manager: audio manager started
2026-04-09T20:14:54.062937Z INFO screenpipe_audio::audio_manager::manager: calendar-assisted speaker diarization: listening for meeting events
2026-04-09T20:14:55.597138Z INFO screenpipe_audio::device::device_manager: starting recording for device: System Audio (output)
2026-04-09T20:14:55.597260Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:14:55.974621Z INFO screenpipe_audio::device::device_manager: starting recording for device: soundcore AeroClip (input)
2026-04-09T20:14:55.974675Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for soundcore AeroClip (input) (30s segments)
2026-04-09T20:15:12.053186Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-206579384513749361, trigger=visual_change)
2026-04-09T20:15:18.079345Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3072302384101791222, trigger=visual_change)
2026-04-09T20:15:25.327517Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-6215291834790472374, trigger=click)
2026-04-09T20:15:28.127631Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 57 eligible frames
2026-04-09T20:15:31.629118Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 56 frames, 25.0MB → 0.9MB (27.5x), 56 JPEGs deleted
2026-04-09T20:16:03.692090Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-4426038861804327400, trigger=visual_change)
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:16:55.543375Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
zsh: terminated npx screenpipe@latest record
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record
detected hardware tier: Mid
checking permissions...
screen recording: ok
microphone: ok
accessibility: ok
2026-04-09T20:18:41.912236Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:18:42.686011Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:18:42.687457Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:18:42.687473Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:18:42.687899Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:18:42.900414Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:18:42.900475Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:18:42.900422Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:18:42.900897Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:18:42.901174Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:18:42.901736Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:18:42.901893Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:18:42.902076Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:18:42.903039Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:18:42.903068Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ false │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ soundcore AeroClip (input) │
│ │ System Audio (output) │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: https://github.com/screenpipe/screenpipe/releases
2026-04-09T20:18:42.906668Z INFO screenpipe: starting UI event capture
2026-04-09T20:18:42.904980Z INFO screenpipe_core::pipes: pipe scheduler started (generation 2)
2026-04-09T2...
|
iTerm2
|
screenpipe"
|
NULL
|
|
ggml_metal_device_init: has bfloat = tr ggml_metal_device_init: has bfloat = true
ggml_metal_device_init: has tensor = false
ggml_metal_device_init: use residency sets = true
ggml_metal_device_init: use shared buffers = true
ggml_metal_device_init: recommendedMaxWorkingSetSize = 11453.25 MB
whisper_init_with_params_no_state: devices = 3
whisper_init_with_params_no_state: backends = 3
whisper_model_load: loading model
whisper_model_load: n_vocab = 51865
whisper_model_load: n_audio_ctx = 1500
whisper_model_load: n_audio_state = 384
whisper_model_load: n_audio_head = 6
whisper_model_load: n_audio_layer = 4
whisper_model_load: n_text_ctx = 448
whisper_model_load: n_text_state = 384
whisper_model_load: n_text_head = 6
whisper_model_load: n_text_layer = 4
whisper_model_load: n_mels = 80
whisper_model_load: ftype = 1
whisper_model_load: qntvr = 0
whisper_model_load: type = 1 (tiny)
whisper_model_load: adding 1608 extra tokens
whisper_model_load: n_langs = 99
whisper_model_load: Metal total size = 77.11 MB
whisper_model_load: model size = 77.11 MB
2026-04-09T20:14:54.059372Z INFO screenpipe_audio::transcription::engine: whisper model loaded successfully
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
2026-04-09T20:14:54.062532Z INFO screenpipe_audio::audio_manager::manager: transcription session created (will be reused across segments)
2026-04-09T20:14:54.062898Z INFO screenpipe_audio::audio_manager::manager: audio manager started
2026-04-09T20:14:54.062937Z INFO screenpipe_audio::audio_manager::manager: calendar-assisted speaker diarization: listening for meeting events
2026-04-09T20:14:55.597138Z INFO screenpipe_audio::device::device_manager: starting recording for device: System Audio (output)
2026-04-09T20:14:55.597260Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for System Audio (output) (30s segments)
2026-04-09T20:14:55.974621Z INFO screenpipe_audio::device::device_manager: starting recording for device: soundcore AeroClip (input)
2026-04-09T20:14:55.974675Z INFO screenpipe_audio::core::run_record_and_transcribe: starting continuous recording for soundcore AeroClip (input) (30s segments)
2026-04-09T20:15:12.053186Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-206579384513749361, trigger=visual_change)
2026-04-09T20:15:18.079345Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=3072302384101791222, trigger=visual_change)
2026-04-09T20:15:25.327517Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-6215291834790472374, trigger=click)
2026-04-09T20:15:28.127631Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 57 eligible frames
2026-04-09T20:15:31.629118Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 56 frames, 25.0MB → 0.9MB (27.5x), 56 JPEGs deleted
2026-04-09T20:16:03.692090Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=-4426038861804327400, trigger=visual_change)
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:16:55.543375Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
zsh: terminated npx screenpipe@latest record
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record
detected hardware tier: Mid
checking permissions...
screen recording: ok
microphone: ok
accessibility: ok
2026-04-09T20:18:41.912236Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:18:42.686011Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:18:42.687457Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:18:42.687473Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:18:42.687899Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:18:42.900414Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:18:42.900475Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:18:42.900422Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:18:42.900897Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:18:42.901174Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:18:42.901736Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:18:42.901893Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:18:42.902076Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:18:42.903039Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:18:42.903068Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ false │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ soundcore AeroClip (input) │
│ │ System Audio (output) │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: https://github.com/screenpipe/screenpipe/releases
2026-04-09T20:18:42.906668Z INFO screenpipe: starting UI event capture
2026-04-09T20:18:42.904980Z INFO screenpipe_core::pipes: pipe scheduler started (generation 2)
2026-04-09T20:18:42.908468Z WARN screenpipe: pi agent install failed: bun not found — install from https://bun.sh
2026-04-09T20:18:42.911739Z INFO screenpipe_engine::p...
|
iTerm2
|
screenpipe"
|
NULL
|
|
whisper_init_state: kv cross size = 9.44 MB
whi whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:16:55.543375Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
zsh: terminated npx screenpipe@latest record
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record
detected hardware tier: Mid
checking permissions...
screen recording: ok
microphone: ok
accessibility: ok
2026-04-09T20:18:41.912236Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:18:42.686011Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:18:42.687457Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:18:42.687473Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:18:42.687899Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:18:42.900414Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:18:42.900475Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:18:42.900422Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:18:42.900897Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:18:42.901174Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:18:42.901736Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:18:42.901893Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:18:42.902076Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:18:42.903039Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:18:42.903068Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ false │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ soundcore AeroClip (input) │
│ │ System Audio (output) │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: https://github.com/screenpipe/screenpipe/releases
2026-04-09T20:18:42.906668Z INFO screenpipe: starting UI event capture
2026-04-09T20:18:42.904980Z INFO screenpipe_core::pipes: pipe scheduler started (generation 2)
2026-04-09T20:18:42.908468Z WARN screenpipe: pi agent install failed: bun not found — install from https://bun.sh
2026-04-09T20:18:42.911739Z INFO screenpipe_engine::power::manager: initial power profile: Performance (on_ac=true, battery=Some(100))
2026-04-09T20:18:42.921315Z INFO screenpipe_engine::ui_recorder: Starting UI event capture
2026-04-09T20:18:42.935181Z INFO screenpipe_engine::ui_recorder: UI recording session started: 9f2b6757-0e2b-45c7-bfae-9666e6db34c5
2026-04-09T20:18:42.935268Z INFO screenpipe_engine::calendar_speaker_id: speaker identification: started (user_name=<not set>)
2026-04-09T20:18:42.935303Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warming from DB (2026-04-08 17:18:42.935300 UTC to 2026-04-09 17:18:42.935300 UTC)
2026-04-09T20:18:42.935997Z INFO screenpipe_engine::meeting_detector: meeting v2: detection loop started (base_interval=5s, profiles=12)
2026-04-09T20:18:42.941704Z INFO screenpipe_engine::server: Server listening on [IP_ADDRESS]:3030
2026-04-09T20:18:42.946772Z INFO screenpipe_connect::mdns: mdns: advertising screenpipe on port 3030
2026-04-09T20:18:42.989922Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warmed with 209 frame entries, coverage from 2026-04-08 17:18:42.935300 UTC
2026-04-09T20:18:43.291350Z INFO screenpipe_engine::vision_manager::manager: Starting vision recording for monitor 1 (1440x900)
2026-04-09T20:18:43.291384Z INFO screenpipe_engine::vision_manager::manager: Starting event-driven capture for monitor 1 (device: monitor_1)
2026-04-09T20:18:43.291397Z INFO screenpipe_engine::vision_manager::manager: Skipping monitor 2 (Display 2_3008x1253_-813,-1253) — not in allowed list
2026-04-09T20:18:43.291406Z INFO screenpipe_engine::vision_manager::monitor_watcher: Starting monitor watcher (polling every 5 seconds)
2026-04-09T20:18:43.291520Z INFO screenpipe_engine::event_driven_capture: event-driven capture started for monitor 1 (device: monitor_1)
2026-04-09T20:18:44.032530Z INFO sck_rs::stream_manager: persistent SCK stream started for display 1 (1440x900, 2fps)
2026-04-09T20:18:47.449731Z INFO screenpipe_engine::event_driven_capture: startup capture for monitor 1: frame_id=210, dur=67ms
2026-04-09T20:18:47.939991Z INFO screenpipe_engine::retention: retention: initialized with 14d retention
2026-04-09T20:18:47.940309Z INFO screenpipe: local retention auto-enabled (14 days)
2026-04-09T20:18:52.910326Z INFO screenpipe_audio::transcription::engine: whisper model available: "/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin"
2026-04-09T20:18:52.910382Z INFO screenpipe_audio::transcription::whisper::model: whisper context: gpu acceleration enabled (Metal on macOS, Vulkan on Windows)
2026-04-09T20:18:52.910389Z INFO screenpipe_audio::transcription::engine: loading whisper model with GPU acceleration...
whisper_init_from_file_with_params_no_state: loading model from '/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin'
whisper_init_with_params_no_state: use gpu = 1
whisper_init_with_params_no_state: flash attn = 0
whisper_init_with_params_no_state: gpu_device = 0
whisper_init_with_params_no_state: dtw = 0
ggml_metal_device_init: tensor API disabled for pre-M5 and pre-A19 devices
ggml_metal_library_init: using embedded metal library
ggml_metal_library_init: loaded in 0.013 sec
ggml_metal_rsets_init: creating a residency set collection (keep_alive = 180 s)
ggml_metal_device_init: GPU name: Apple M1
ggml_metal_device_init: GPU family: MTLGPUFamilyApple7 (1007)
ggml_metal_device_init: GPU family: MTLGPUFamilyCommon3 (3003)
ggml_metal_device_init: GPU family: MTLGPUFamilyMetal3 (5001)
ggml_metal_device_init: simdgroup reduction = true
ggml_metal_device_init: simdgroup matrix mul. = true
ggml_metal_device_init: has unified memory = true
ggml_metal_device_init: has bfloat = true
ggml_metal_device_init: has tensor = false
ggml_metal_device_init: use residency sets = true
ggml_metal_device_init: use shared buffers = true
ggml_metal_device_init: recommendedMaxWorkingSetSize = 11453.25 MB
whisper_init_with_params_no_state: devices = 3
whisper_init_with_params_no_state: backends = 3
whisper_model_load: loading model
whisper_model_load: n_vocab = 51865
whisper_model_load: n_audio_ctx = 1500
whisper_model_load: n_audio_state = 384
whisper_model_load: n_audio_head = 6
whisper_model_load: n_audio_layer = 4
whisper_model_load: n_text_ctx = 448
whisper...
|
iTerm2
|
screenpipe"
|
NULL
|
|
whisper_init_state: compute buffer (cross) = 8 whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:16:55.543375Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
zsh: terminated npx screenpipe@latest record
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record
detected hardware tier: Mid
checking permissions...
screen recording: ok
microphone: ok
accessibility: ok
2026-04-09T20:18:41.912236Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:18:42.686011Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:18:42.687457Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:18:42.687473Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:18:42.687899Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:18:42.900414Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:18:42.900475Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:18:42.900422Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:18:42.900897Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:18:42.901174Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:18:42.901736Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:18:42.901893Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:18:42.902076Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:18:42.903039Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:18:42.903068Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ false │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ soundcore AeroClip (input) │
│ │ System Audio (output) │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: https://github.com/screenpipe/screenpipe/releases
2026-04-09T20:18:42.906668Z INFO screenpipe: starting UI event capture
2026-04-09T20:18:42.904980Z INFO screenpipe_core::pipes: pipe scheduler started (generation 2)
2026-04-09T20:18:42.908468Z WARN screenpipe: pi agent install failed: bun not found — install from https://bun.sh
2026-04-09T20:18:42.911739Z INFO screenpipe_engine::power::manager: initial power profile: Performance (on_ac=true, battery=Some(100))
2026-04-09T20:18:42.921315Z INFO screenpipe_engine::ui_recorder: Starting UI event capture
2026-04-09T20:18:42.935181Z INFO screenpipe_engine::ui_recorder: UI recording session started: 9f2b6757-0e2b-45c7-bfae-9666e6db34c5
2026-04-09T20:18:42.935268Z INFO screenpipe_engine::calendar_speaker_id: speaker identification: started (user_name=<not set>)
2026-04-09T20:18:42.935303Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warming from DB (2026-04-08 17:18:42.935300 UTC to 2026-04-09 17:18:42.935300 UTC)
2026-04-09T20:18:42.935997Z INFO screenpipe_engine::meeting_detector: meeting v2: detection loop started (base_interval=5s, profiles=12)
2026-04-09T20:18:42.941704Z INFO screenpipe_engine::server: Server listening on [IP_ADDRESS]:3030
2026-04-09T20:18:42.946772Z INFO screenpipe_connect::mdns: mdns: advertising screenpipe on port 3030
2026-04-09T20:18:42.989922Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warmed with 209 frame entries, coverage from 2026-04-08 17:18:42.935300 UTC
2026-04-09T20:18:43.291350Z INFO screenpipe_engine::vision_manager::manager: Starting vision recording for monitor 1 (1440x900)
2026-04-09T20:18:43.291384Z INFO screenpipe_engine::vision_manager::manager: Starting event-driven capture for monitor 1 (device: monitor_1)
2026-04-09T20:18:43.291397Z INFO screenpipe_engine::vision_manager::manager: Skipping monitor 2 (Display 2_3008x1253_-813,-1253) — not in allowed list
2026-04-09T20:18:43.291406Z INFO screenpipe_engine::vision_manager::monitor_watcher: Starting monitor watcher (polling every 5 seconds)
2026-04-09T20:18:43.291520Z INFO screenpipe_engine::event_driven_capture: event-driven capture started for monitor 1 (device: monitor_1)
2026-04-09T20:18:44.032530Z INFO sck_rs::stream_manager: persistent SCK stream started for display 1 (1440x900, 2fps)
2026-04-09T20:18:47.449731Z INFO screenpipe_engine::event_driven_capture: startup capture for monitor 1: frame_id=210, dur=67ms
2026-04-09T20:18:47.939991Z INFO screenpipe_engine::retention: retention: initialized with 14d retention
2026-04-09T20:18:47.940309Z INFO screenpipe: local retention auto-enabled (14 days)
2026-04-09T20:18:52.910326Z INFO screenpipe_audio::transcription::engine: whisper model available: "/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin"
2026-04-09T20:18:52.910382Z INFO screenpipe_audio::transcription::whisper::model: whisper context: gpu acceleration enabled (Metal on macOS, Vulkan on Windows)
2026-04-09T20:18:52.910389Z INFO screenpipe_audio::transcription::engine: loading whisper model with GPU acceleration...
whisper_init_from_file_with_params_no_state: loading model from '/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin'
whisper_init_with_params_no_state: use gpu = 1
whisper_init_with_params_no_state: flash attn = 0
whisper_init_with_params_no_state: gpu_device = 0
whisper_init_with_params_no_state: dtw = 0
ggml_metal_device_init: tensor API disabled for pre-M5 and pre-A19 devices
ggml_metal_library_init: using embedded metal library
ggml_metal_library_init: loaded in 0.013 sec
ggml_metal_rsets_init: creating a residency set collection (keep_alive = 180 s)
ggml_metal_device_init: GPU name: Apple M1
ggml_metal_device_init: GPU family: MTLGPUFamilyApple7 (1007)
ggml_metal_device_init: GPU family: MTLGPUFamilyCommon3 (3003)
ggml_metal_device_init: GPU family: MTLGPUFamilyMetal3 (5001)
ggml_metal_device_init: simdgroup reduction = true
ggml_metal_device_init: simdgroup matrix mul. = true
ggml_metal_device_init: has unified memory = true
ggml_metal_device_init: has bfloat = true
ggml_metal_device_init: has tensor = false
ggml_metal_device_init: use residency sets = true
ggml_metal_device_init: use shared buffers = true
ggml_metal_device_init: recommendedMaxWorkingSetSize = 11453.25 MB
whisper_init_with_params_no_state: devices = 3
whisper_init_with_params_no_state: backends = 3
whisper_model_load: loading model
whisper_model_load: n_vocab = 51865
whisper_model_load: n_audio_ctx = 1500
whisper_model_load: n_audio_state = 384
whisper_model_load: n_audio_head = 6
whisper_model_load: n_audio_layer = 4
whisper_model_load: n_text_ctx = 448
whisper_model_load: n_text_state = 384
whisper_model_load: n_text_head = 6
whisper_model_load: n_text_layer = 4
whisper_model_load: n_mels = 80
whisper_model_load: ftype = 1
whisper_model_load: qn...
|
iTerm2
|
screenpipe"
|
NULL
|
|
whisper_init_state: compute buffer (decode) = 96 whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:16:55.543375Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
zsh: terminated npx screenpipe@latest record
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record
detected hardware tier: Mid
checking permissions...
screen recording: ok
microphone: ok
accessibility: ok
2026-04-09T20:18:41.912236Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:18:42.686011Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:18:42.687457Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:18:42.687473Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:18:42.687899Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:18:42.900414Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:18:42.900475Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:18:42.900422Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:18:42.900897Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:18:42.901174Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:18:42.901736Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:18:42.901893Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:18:42.902076Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:18:42.903039Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:18:42.903068Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ false │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ soundcore AeroClip (input) │
│ │ System Audio (output) │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: https://github.com/screenpipe/screenpipe/releases
2026-04-09T20:18:42.906668Z INFO screenpipe: starting UI event capture
2026-04-09T20:18:42.904980Z INFO screenpipe_core::pipes: pipe scheduler started (generation 2)
2026-04-09T20:18:42.908468Z WARN screenpipe: pi agent install failed: bun not found — install from https://bun.sh
2026-04-09T20:18:42.911739Z INFO screenpipe_engine::power::manager: initial power profile: Performance (on_ac=true, battery=Some(100))
2026-04-09T20:18:42.921315Z INFO screenpipe_engine::ui_recorder: Starting UI event capture
2026-04-09T20:18:42.935181Z INFO screenpipe_engine::ui_recorder: UI recording session started: 9f2b6757-0e2b-45c7-bfae-9666e6db34c5
2026-04-09T20:18:42.935268Z INFO screenpipe_engine::calendar_speaker_id: speaker identification: started (user_name=<not set>)
2026-04-09T20:18:42.935303Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warming from DB (2026-04-08 17:18:42.935300 UTC to 2026-04-09 17:18:42.935300 UTC)
2026-04-09T20:18:42.935997Z INFO screenpipe_engine::meeting_detector: meeting v2: detection loop started (base_interval=5s, profiles=12)
2026-04-09T20:18:42.941704Z INFO screenpipe_engine::server: Server listening on [IP_ADDRESS]:3030
2026-04-09T20:18:42.946772Z INFO screenpipe_connect::mdns: mdns: advertising screenpipe on port 3030
2026-04-09T20:18:42.989922Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warmed with 209 frame entries, coverage from 2026-04-08 17:18:42.935300 UTC
2026-04-09T20:18:43.291350Z INFO screenpipe_engine::vision_manager::manager: Starting vision recording for monitor 1 (1440x900)
2026-04-09T20:18:43.291384Z INFO screenpipe_engine::vision_manager::manager: Starting event-driven capture for monitor 1 (device: monitor_1)
2026-04-09T20:18:43.291397Z INFO screenpipe_engine::vision_manager::manager: Skipping monitor 2 (Display 2_3008x1253_-813,-1253) — not in allowed list
2026-04-09T20:18:43.291406Z INFO screenpipe_engine::vision_manager::monitor_watcher: Starting monitor watcher (polling every 5 seconds)
2026-04-09T20:18:43.291520Z INFO screenpipe_engine::event_driven_capture: event-driven capture started for monitor 1 (device: monitor_1)
2026-04-09T20:18:44.032530Z INFO sck_rs::stream_manager: persistent SCK stream started for display 1 (1440x900, 2fps)
2026-04-09T20:18:47.449731Z INFO screenpipe_engine::event_driven_capture: startup capture for monitor 1: frame_id=210, dur=67ms
2026-04-09T20:18:47.939991Z INFO screenpipe_engine::retention: retention: initialized with 14d retention
2026-04-09T20:18:47.940309Z INFO screenpipe: local retention auto-enabled (14 days)
2026-04-09T20:18:52.910326Z INFO screenpipe_audio::transcription::engine: whisper model available: "/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin"
2026-04-09T20:18:52.910382Z INFO screenpipe_audio::transcription::whisper::model: whisper context: gpu acceleration enabled (Metal on macOS, Vulkan on Windows)
2026-04-09T20:18:52.910389Z INFO screenpipe_audio::transcription::engine: loading whisper model with GPU acceleration...
whisper_init_from_file_with_params_no_state: loading model from '/Users/lukas/.cache/huggingface/hub/models--ggerganov--whisper.cpp/snapshots/5359861c739e955e79d9a303bcbc70fb988958b1/ggml-tiny.bin'
whisper_init_with_params_no_state: use gpu = 1
whisper_init_with_params_no_state: flash attn = 0
whisper_init_with_params_no_state: gpu_device = 0
whisper_init_with_params_no_state: dtw = 0
ggml_metal_device_init: tensor API disabled for pre-M5 and pre-A19 devices
ggml_metal_library_init: using embedded metal library
ggml_metal_library_init: loaded in 0.013 sec
ggml_metal_rsets_init: creating a residency set collection (keep_alive = 180 s)
ggml_metal_device_init: GPU name: Apple M1
ggml_metal_device_init: GPU family: MTLGPUFamilyApple7 (1007)
ggml_metal_device_init: GPU family: MTLGPUFamilyCommon3 (3003)
ggml_metal_device_init: GPU family: MTLGPUFamilyMetal3 (5001)
ggml_metal_device_init: simdgroup reduction = true
ggml_metal_device_init: simdgroup matrix mul. = true
ggml_metal_device_init: has unified memory = true
ggml_metal_device_init: has bfloat = true
ggml_metal_device_init: has tensor = false
ggml_metal_device_init: use residency sets = true
ggml_metal_device_init: use shared buffers = true
ggml_metal_device_init: recommendedMaxWorkingSetSize = 11453.25 MB
whisper_init_with_params_no_state: devices = 3
whisper_init_with_params_no_state: backends = 3
whisper_model_load: loading model
whisper_model_load: n_vocab = 51865
whisper_model_load: n_audio_ctx = 1500
whisper_model_load: n_audio_state = 384
whisper_model_load: n_audio_head = 6
whisper_model_load: n_audio_layer = 4
whisper_model_load: n_text_ctx = 448
whisper_model_load: n_text_state = 384
whisper_model_load: n_text_head = 6
whisper_model_load: n_text_layer = 4
whisper_model_load: n_mels = 80
whisper_model_load: ftype = 1
whisper_model_load: qntvr = 0
whisper_model_load: type = 1 (ti...
|
iTerm2
|
screenpipe"
|
NULL
|
|
UW PICO 5.09 UW PICO 5.09 New Buffer
[ Read 3 lines ]
^G Get Help ^O WriteOut ^R Read File ^Y Prev Pg ^K Cut Text ^C Cur Pos
^X Exit ^J Justify ^W Where is ^V Next Pg ^U UnCut Text ^T To Spell
Last login: Thu Apr 9 19:54:42 on ttys010
UW PICO 5.09 New Buffer
[ Read 11 lines ]
^G Get Help ^O WriteOut ^R Read File ^Y Prev Pg ^K Cut Text ^C Cur Pos
^X Exit ^J Justify ^W Where is ^V Next Pg ^U UnCut Text ^T To Spell
Poetry could not find a pyproject.toml file in /Users/lukas or its parents
Poetry could not find a pyproject.toml file in /Users/lukas or its parents
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ sqlite3 ~/.screenpipe/db.sqlite "SELECT app_name, window_name, timestamp FROM frames ORDER BY timestamp DESC LIMIT 10;"
iTerm2|-zsh|2026-04-09T17:09:56.277988+00:00
||2026-04-09T17:09:54.152728+00:00
iTerm2|✳ Check today's app usage|2026-04-09T17:09:52.277880+00:00
||2026-04-09T17:09:50.551186+00:00
||2026-04-09T17:09:48.080710+00:00
||2026-04-09T17:09:44.554361+00:00
||2026-04-09T17:09:42.248132+00:00
||2026-04-09T17:09:38.373309+00:00
iTerm2|✳ Check today's app usage|2026-04-09T17:09:23.768571+00:00
iTerm2|✳ Check today's app usage|2026-04-09T17:09:12.596384+00:00
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ cd ~/.screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ ll
total 55496
drwxr-xr-x 8 lukas staff 256 9 Apr 19:53 .
drwx------+ 88 lukas staff 2816 9 Apr 20:09 ..
drwxr-xr-x 74 lukas staff 2368 9 Apr 20:10 data
-rw-r--r-- 1 lukas staff 15765504 9 Apr 20:08 db.sqlite
-rw-r--r-- 1 lukas staff 32768 9 Apr 19:53 db.sqlite-shm
-rw-r--r-- 1 lukas staff 11696712 9 Apr 20:11 db.sqlite-wal
drwxr-xr-x 8 lukas staff 256 9 Apr 19:53 pipes
-rw-r--r-- 1 lukas staff 25189 9 Apr 20:09 screenpipe.2026-04-09.0.log
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ cat > ~/.screenpipe/config.json << 'EOF'
{
"disable_audio": true
}
EOF
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ ll
total 55504
drwxr-xr-x 9 lukas staff 288 9 Apr 20:11 .
drwx------+ 88 lukas staff 2816 9 Apr 20:09 ..
-rw-r--r-- 1 lukas staff 28 9 Apr 20:11 config.json
drwxr-xr-x 74 lukas staff 2368 9 Apr 20:10 data
-rw-r--r-- 1 lukas staff 15765504 9 Apr 20:08 db.sqlite
-rw-r--r-- 1 lukas staff 32768 9 Apr 19:53 db.sqlite-shm
-rw-r--r-- 1 lukas staff 12195232 9 Apr 20:11 db.sqlite-wal
drwxr-xr-x 8 lukas staff 256 9 Apr 19:53 pipes
-rw-r--r-- 1 lukas staff 25189 9 Apr 20:09 screenpipe.2026-04-09.0.log
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ nano config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ nano config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ ls -la ~/.screenpipe/config.json
-rw-r--r-- 1 lukas staff 177 9 Apr 20:13 /Users/lukas/.screenpipe/config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ pkill -f screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ npx screenpipe@latest record --disable-audio
detected hardware tier: Mid
checking permissions...
screen recording: ok
accessibility: ok
2026-04-09T20:19:27.531337Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:19:28.280889Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:19:28.282336Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:19:28.282354Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:19:28.282617Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:19:28.282705Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:19:28.282660Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:19:28.283417Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:19:28.283684Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:19:28.284452Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:19:28.284646Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:19:28.284821Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:19:28.285013Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:19:28.285763Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:19:28.285784Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ true │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ disabled │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: https://github.com/screenpipe/screenpipe/releases
2026-04-09T20:19:28.287964Z INFO screenpipe: starting UI event capture
2026-04-09T20:19:28.287733Z INFO screenpipe_core::pipes: pipe scheduler started (generation 2)
2026-04-09T20:19:28.292125Z WARN screenpipe: pi agent install failed: bun not found — install from https://bun.sh
2026-04-09T20:19:28.294471Z INFO screenpipe_engine::power::manager: initial power profile: Performance (on_ac=true, battery=Some(100))
2026-04-09T20:19:28.305266Z INFO screenpipe_engine::ui_recorder: Starting UI event capture
2026-04-09T20:19:28.320845Z INFO screenpipe_engine::ui_recorder: UI recording session started: 7e197a7d-7f33-4f4b-ba21-2fb99f1259cf
2026-04-09T20:19:28.320921Z INFO screenpipe_engine::calendar_speaker_id: speaker identification: started (user_name=<not set>)
2026-04-09T20:19:28.320981Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warming from DB (2026-04-08 17:19:28.320980 UTC to 2026-04-09 17:19:28.320980 UTC)
2026-04-09T20:19:28.321784Z INFO screenpipe_engine::meeting_detector: meeting v2: detection loop started (base_interval=5s, profiles=12)
2026-04-09T20:19:28.325794Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warmed with 222 frame entries, coverage from 2026-04-08 17:19:28.320980 UTC
2026-04-09T20:19:28.328661Z ERROR screenpipe: server stopped with error: Os { code: 48, kind: AddrInUse, message: "Address already in use" }
2026-04-09T20:19:28.733056Z INFO screenpipe_engine::vision_manager::manager: Starting vision recording for monitor 1 (1440x900)
2026-04-09T20:19:28.733097Z INFO screenpipe_engine::vision_manager::manager: Starting event-driven capture for monitor 1 (device: monitor_1)
2026-04-09T20:19:28.733118Z INFO screenpipe_engine::vision_manager::manager: Skipping monitor 2 (Display 2_3008x1253_-813,-1253) — not in allowed list
2026-04-09T20:19:28.733132Z INFO screenpipe_engine::vision_manager::monitor_watcher: Starting monitor watcher (polling every 5 seconds)
2026-04-09T20:19:28.733186Z INFO screenpipe_engine::event_driven_capture: event-driven capture started for monitor 1 (device: monitor_1)
DOCKER
Close Tab
DEV (-zsh)
Close Tab
APP (-zsh)
Close Tab
✳ Check today's app usage (claude)
Close Tab
screenpipe"
Close Tab
screenpipe"
Close Tab
⌥⌘1
screenpipe"...
|
iTerm2
|
screenpipe"
|
NULL
|
|
UW PICO 5.09 UW PICO 5.09 New Buffer
[ Read 3 lines ]
^G Get Help ^O WriteOut ^R Read File ^Y Prev Pg ^K Cut Text ^C Cur Pos
^X Exit ^J Justify ^W Where is ^V Next Pg ^U UnCut Text ^T To Spell
Last login: Thu Apr 9 19:54:42 on ttys010
UW PICO 5.09 New Buffer
[ Read 11 lines ]
^G Get Help ^O WriteOut ^R Read File ^Y Prev Pg ^K Cut Text ^C Cur Pos
^X Exit ^J Justify ^W Where is ^V Next Pg ^U UnCut Text ^T To Spell
Poetry could not find a pyproject.toml file in /Users/lukas or its parents
Poetry could not find a pyproject.toml file in /Users/lukas or its parents
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ sqlite3 ~/.screenpipe/db.sqlite "SELECT app_name, window_name, timestamp FROM frames ORDER BY timestamp DESC LIMIT 10;"
iTerm2|-zsh|2026-04-09T17:09:56.277988+00:00
||2026-04-09T17:09:54.152728+00:00
iTerm2|✳ Check today's app usage|2026-04-09T17:09:52.277880+00:00
||2026-04-09T17:09:50.551186+00:00
||2026-04-09T17:09:48.080710+00:00
||2026-04-09T17:09:44.554361+00:00
||2026-04-09T17:09:42.248132+00:00
||2026-04-09T17:09:38.373309+00:00
iTerm2|✳ Check today's app usage|2026-04-09T17:09:23.768571+00:00
iTerm2|✳ Check today's app usage|2026-04-09T17:09:12.596384+00:00
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ cd ~/.screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ ll
total 55496
drwxr-xr-x 8 lukas staff 256 9 Apr 19:53 .
drwx------+ 88 lukas staff 2816 9 Apr 20:09 ..
drwxr-xr-x 74 lukas staff 2368 9 Apr 20:10 data
-rw-r--r-- 1 lukas staff 15765504 9 Apr 20:08 db.sqlite
-rw-r--r-- 1 lukas staff 32768 9 Apr 19:53 db.sqlite-shm
-rw-r--r-- 1 lukas staff 11696712 9 Apr 20:11 db.sqlite-wal
drwxr-xr-x 8 lukas staff 256 9 Apr 19:53 pipes
-rw-r--r-- 1 lukas staff 25189 9 Apr 20:09 screenpipe.2026-04-09.0.log
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ cat > ~/.screenpipe/config.json << 'EOF'
{
"disable_audio": true
}
EOF
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ ll
total 55504
drwxr-xr-x 9 lukas staff 288 9 Apr 20:11 .
drwx------+ 88 lukas staff 2816 9 Apr 20:09 ..
-rw-r--r-- 1 lukas staff 28 9 Apr 20:11 config.json
drwxr-xr-x 74 lukas staff 2368 9 Apr 20:10 data
-rw-r--r-- 1 lukas staff 15765504 9 Apr 20:08 db.sqlite
-rw-r--r-- 1 lukas staff 32768 9 Apr 19:53 db.sqlite-shm
-rw-r--r-- 1 lukas staff 12195232 9 Apr 20:11 db.sqlite-wal
drwxr-xr-x 8 lukas staff 256 9 Apr 19:53 pipes
-rw-r--r-- 1 lukas staff 25189 9 Apr 20:09 screenpipe.2026-04-09.0.log
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ nano config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ nano config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ ls -la ~/.screenpipe/config.json
-rw-r--r-- 1 lukas staff 177 9 Apr 20:13 /Users/lukas/.screenpipe/config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ pkill -f screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ npx screenpipe@latest record --disable-audio
detected hardware tier: Mid
checking permissions...
screen recording: ok
accessibility: ok
2026-04-09T20:19:27.531337Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:19:28.280889Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:19:28.282336Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:19:28.282354Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:19:28.282617Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:19:28.282705Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:19:28.282660Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:19:28.283417Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:19:28.283684Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:19:28.284452Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:19:28.284646Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:19:28.284821Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:19:28.285013Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:19:28.285763Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:19:28.285784Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ true │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ disabled │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: https://github.com/screenpipe/screenpipe/releases
2026-04-09T20:19:28.287964Z INFO screenpipe: starting UI event capture
2026-04-09T20:19:28.287733Z INFO screenpipe_core::pipes: pipe scheduler started (generation 2)
2026-04-09T20:19:28.292125Z WARN screenpipe: pi agent install failed: bun not found — install from https://bun.sh
2026-04-09T20:19:28.294471Z INFO screenpipe_engine::power::manager: initial power profile: Performance (on_ac=true, battery=Some(100))
2026-04-09T20:19:28.305266Z INFO screenpipe_engine::ui_recorder: Starting UI event capture
2026-04-09T20:19:28.320845Z INFO screenpipe_engine::ui_recorder: UI recording session started: 7e197a7d-7f33-4f4b-ba21-2fb99f1259cf
2026-04-09T20:19:28.320921Z INFO screenpipe_engine::calendar_speaker_id: speaker identification: started (user_name=<not set>)
2026-04-09T20:19:28.320981Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warming from DB (2026-04-08 17:19:28.320980 UTC to 2026-04-09 17:19:28.320980 UTC)
2026-04-09T20:19:28.321784Z INFO screenpipe_engine::meeting_detector: meeting v2: detection loop started (base_interval=5s, profiles=12)
2026-04-09T20:19:28.325794Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warmed with 222 frame entries, coverage from 2026-04-08 17:19:28.320980 UTC
2026-04-09T20:19:28.328661Z ERROR screenpipe: server stopped with error: Os { code: 48, kind: AddrInUse, message: "Address already in use" }
2026-04-09T20:19:28.733056Z INFO screenpipe_engine::vision_manager::manager: Starting vision recording for monitor 1 (1440x900)
2026-04-09T20:19:28.733097Z INFO screenpipe_engine::vision_manager::manager: Starting event-driven capture for monitor 1 (device: monitor_1)
2026-04-09T20:19:28.733118Z INFO screenpipe_engine::vision_manager::manager: Skipping monitor 2 (Display 2_3008x1253_-813,-1253) — not in allowed list
2026-04-09T20:19:28.733132Z INFO screenpipe_engine::vision_manager::monitor_watcher: Starting monitor watcher (polling every 5 seconds)
2026-04-09T20:19:28.733186Z INFO screenpipe_engine::event_driven_capture: event-driven capture started for monitor 1 (device: monitor_1)
2026-04-09T20:19:29.536587Z INFO sck_rs::stream_manager: persistent SCK stream started for display 1 (1440x900, 2fps)
2026-04-09T20:19:29.615611Z INFO screenpipe_engine::event_driven_capture: startup capture for monitor 1: frame_id=223, dur=33ms
DOCKER
Close Tab
DEV (-zsh)
Close Tab
APP (-zsh)
Close Tab
✳ Check today's app usage (claude)
Close Tab
screenpipe"
Close Tab
screenpipe"
Close Tab
⌥⌘1
screenpipe"...
|
iTerm2
|
screenpipe"
|
NULL
|
|
whisper_backend_init_gpu: device 0: Metal (type: 1 whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:16:55.543375Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
zsh: terminated npx screenpipe@latest record
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record
detected hardware tier: Mid
checking permissions...
screen recording: ok
microphone: ok
accessibility: ok
2026-04-09T20:18:41.912236Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:18:42.686011Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:18:42.687457Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:18:42.687473Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:18:42.687899Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:18:42.900414Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:18:42.900475Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:18:42.900422Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:18:42.900897Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:18:42.901174Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:18:42.901736Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:18:42.901893Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:18:42.902076Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:18:42.903039Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:18:42.903068Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ false │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ soundcore AeroClip (input) │
│ │ System Audio (output) │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: [URL_WITH_CREDENTIALS] record
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record --disable-audio
detected hardware tier: Mid
checking permissions...
screen recording: ok
accessibility: ok
2026-04-09T20:19:52.412195Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:19:53.055756Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:19:53.058756Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:19:53.059017Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:19:53.059054Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:19:53.058846Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:19:53.059166Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:19:53.059575Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:19:53.059559Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:19:53.059832Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:19:53.060406Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:19:53.060587Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:19:53.060779Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:19:53.061707Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:19:53.061733Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ ...
|
iTerm2
|
screenpipe"
|
NULL
|
|
whisper_backend_init_gpu: found GPU device 0: Meta whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:16:55.543375Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
zsh: terminated npx screenpipe@latest record
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record
detected hardware tier: Mid
checking permissions...
screen recording: ok
microphone: ok
accessibility: ok
2026-04-09T20:18:41.912236Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:18:42.686011Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:18:42.687457Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:18:42.687473Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:18:42.687899Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:18:42.900414Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:18:42.900475Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:18:42.900422Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:18:42.900897Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:18:42.901174Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:18:42.901736Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:18:42.901893Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:18:42.902076Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:18:42.903039Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:18:42.903068Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ false │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ soundcore AeroClip (input) │
│ │ System Audio (output) │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: [URL_WITH_CREDENTIALS] record
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record --disable-audio
detected hardware tier: Mid
checking permissions...
screen recording: ok
accessibility: ok
2026-04-09T20:19:52.412195Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:19:53.055756Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:19:53.058756Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:19:53.059017Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:19:53.059054Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:19:53.058846Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:19:53.059166Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:19:53.059575Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:19:53.059559Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:19:53.059832Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:19:53.060406Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:19:53.060587Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:19:53.060779Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:19:53.061707Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:19:53.061733Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ ...
|
iTerm2
|
screenpipe"
|
NULL
|
|
ggml_metal_init: allocating
ggml_metal_init: found ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:16:55.543375Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
zsh: terminated npx screenpipe@latest record
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record
detected hardware tier: Mid
checking permissions...
screen recording: ok
microphone: ok
accessibility: ok
2026-04-09T20:18:41.912236Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:18:42.686011Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:18:42.687457Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:18:42.687473Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:18:42.687899Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:18:42.900414Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:18:42.900475Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:18:42.900422Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:18:42.900897Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:18:42.901174Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:18:42.901736Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:18:42.901893Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:18:42.902076Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:18:42.903039Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:18:42.903068Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ false │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ soundcore AeroClip (input) │
│ │ System Audio (output) │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: [URL_WITH_CREDENTIALS] record
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record --disable-audio
detected hardware tier: Mid
checking permissions...
screen recording: ok
accessibility: ok
2026-04-09T20:19:52.412195Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:19:53.055756Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:19:53.058756Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:19:53.059017Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:19:53.059054Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:19:53.058846Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:19:53.059166Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:19:53.059575Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:19:53.059559Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:19:53.059832Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:19:53.060406Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:19:53.060587Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:19:53.060779Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:19:53.061707Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:19:53.061733Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
...
|
iTerm2
|
screenpipe"
|
NULL
|
|
ggml_metal_init: found device: Apple M1
ggml_metal ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:16:55.543375Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
zsh: terminated npx screenpipe@latest record
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record
detected hardware tier: Mid
checking permissions...
screen recording: ok
microphone: ok
accessibility: ok
2026-04-09T20:18:41.912236Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:18:42.686011Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:18:42.687457Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:18:42.687473Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:18:42.687899Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:18:42.900414Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:18:42.900475Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:18:42.900422Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:18:42.900897Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:18:42.901174Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:18:42.901736Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:18:42.901893Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:18:42.902076Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:18:42.903039Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:18:42.903068Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ false │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ soundcore AeroClip (input) │
│ │ System Audio (output) │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: [URL_WITH_CREDENTIALS] record
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record --disable-audio
detected hardware tier: Mid
checking permissions...
screen recording: ok
accessibility: ok
2026-04-09T20:19:52.412195Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:19:53.055756Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:19:53.058756Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:19:53.059017Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:19:53.059054Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:19:53.058846Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:19:53.059166Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:19:53.059575Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:19:53.059559Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:19:53.059832Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:19:53.060406Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:19:53.060587Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:19:53.060779Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:19:53.061707Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:19:53.061733Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
p...
|
iTerm2
|
screenpipe"
|
NULL
|
|
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:16:55.543375Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
zsh: terminated npx screenpipe@latest record
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record
detected hardware tier: Mid
checking permissions...
screen recording: ok
microphone: ok
accessibility: ok
2026-04-09T20:18:41.912236Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:18:42.686011Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:18:42.687457Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:18:42.687473Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:18:42.687899Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:18:42.900414Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:18:42.900475Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:18:42.900422Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:18:42.900897Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:18:42.901174Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:18:42.901736Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:18:42.901893Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:18:42.902076Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:18:42.903039Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:18:42.903068Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ false │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ soundcore AeroClip (input) │
│ │ System Audio (output) │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: [URL_WITH_CREDENTIALS] record
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record --disable-audio
detected hardware tier: Mid
checking permissions...
screen recording: ok
accessibility: ok
2026-04-09T20:19:52.412195Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:19:53.055756Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:19:53.058756Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:19:53.059017Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:19:53.059054Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:19:53.058846Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:19:53.059166Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:19:53.059575Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:19:53.059559Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:19:53.059832Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:19:53.060406Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:19:53.060587Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:19:53.060779Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:19:53.061707Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:19:53.061733Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said ...
|
iTerm2
|
screenpipe"
|
NULL
|
|
ggml_metal_init: use fusion = true
ggml_me ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:16:55.543375Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
zsh: terminated npx screenpipe@latest record
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record
detected hardware tier: Mid
checking permissions...
screen recording: ok
microphone: ok
accessibility: ok
2026-04-09T20:18:41.912236Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:18:42.686011Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:18:42.687457Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:18:42.687473Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:18:42.687899Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:18:42.900414Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:18:42.900475Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:18:42.900422Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:18:42.900897Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:18:42.901174Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:18:42.901736Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:18:42.901893Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:18:42.902076Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:18:42.903039Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:18:42.903068Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ false │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ soundcore AeroClip (input) │
│ │ System Audio (output) │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: [URL_WITH_CREDENTIALS] record
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record --disable-audio
detected hardware tier: Mid
checking permissions...
screen recording: ok
accessibility: ok
2026-04-09T20:19:52.412195Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:19:53.055756Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:19:53.058756Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:19:53.059017Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:19:53.059054Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:19:53.058846Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:19:53.059166Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:19:53.059575Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:19:53.059559Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:19:53.059832Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:19:53.060406Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:19:53.060587Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:19:53.060779Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:19:53.061707Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:19:53.061733Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer fr...
|
iTerm2
|
screenpipe"
|
NULL
|
|
ggml_metal_init: use graph optimize = true
whisper ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
whisper_backend_init_gpu: device 0: Metal (type: 1)
whisper_backend_init_gpu: found GPU device 0: Metal (type: 1, cnt: 0)
whisper_backend_init_gpu: using Metal backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M1
ggml_metal_init: picking default device: Apple M1
ggml_metal_init: use fusion = true
ggml_metal_init: use concurrency = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size = 3.15 MB
whisper_init_state: kv cross size = 9.44 MB
whisper_init_state: kv pad size = 2.36 MB
whisper_init_state: compute buffer (conv) = 14.17 MB
whisper_init_state: compute buffer (encode) = 65.96 MB
whisper_init_state: compute buffer (cross) = 8.50 MB
whisper_init_state: compute buffer (decode) = 96.83 MB
ggml_metal_free: deallocating
2026-04-09T20:16:55.543375Z INFO screenpipe_audio::audio_manager::manager: reconciliation: transcribed 50 orphaned chunks
zsh: terminated npx screenpipe@latest record
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record
detected hardware tier: Mid
checking permissions...
screen recording: ok
microphone: ok
accessibility: ok
2026-04-09T20:18:41.912236Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:18:42.686011Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:18:42.687457Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:18:42.687473Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:18:42.687899Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:18:42.900414Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:18:42.900475Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:18:42.900422Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:18:42.900897Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:18:42.901174Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:18:42.901736Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:18:42.901893Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:18:42.902076Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:18:42.903039Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:18:42.903068Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ false │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ soundcore AeroClip (input) │
│ │ System Audio (output) │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: [URL_WITH_CREDENTIALS] record
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ npx screenpipe@latest record --disable-audio
detected hardware tier: Mid
checking permissions...
screen recording: ok
accessibility: ok
2026-04-09T20:19:52.412195Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:19:53.055756Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:19:53.058756Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:19:53.059017Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:19:53.059054Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:19:53.058846Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:19:53.059166Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:19:53.059575Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:19:53.059559Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:19:53.059832Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:19:53.060406Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:19:53.060587Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:19:53.060779Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:19:53.061707Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:19:53.061733Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────...
|
iTerm2
|
screenpipe"
|
NULL
|
|
UW PICO 5.09 UW PICO 5.09 New Buffer
[ Read 3 lines ]
^G Get Help ^O WriteOut ^R Read File ^Y Prev Pg ^K Cut Text ^C Cur Pos
^X Exit ^J Justify ^W Where is ^V Next Pg ^U UnCut Text ^T To Spell
Last login: Thu Apr 9 19:54:42 on ttys010
UW PICO 5.09 New Buffer
[ Read 11 lines ]
^G Get Help ^O WriteOut ^R Read File ^Y Prev Pg ^K Cut Text ^C Cur Pos
^X Exit ^J Justify ^W Where is ^V Next Pg ^U UnCut Text ^T To Spell
Poetry could not find a pyproject.toml file in /Users/lukas or its parents
Poetry could not find a pyproject.toml file in /Users/lukas or its parents
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ sqlite3 ~/.screenpipe/db.sqlite "SELECT app_name, window_name, timestamp FROM frames ORDER BY timestamp DESC LIMIT 10;"
iTerm2|-zsh|2026-04-09T17:09:56.277988+00:00
||2026-04-09T17:09:54.152728+00:00
iTerm2|✳ Check today's app usage|2026-04-09T17:09:52.277880+00:00
||2026-04-09T17:09:50.551186+00:00
||2026-04-09T17:09:48.080710+00:00
||2026-04-09T17:09:44.554361+00:00
||2026-04-09T17:09:42.248132+00:00
||2026-04-09T17:09:38.373309+00:00
iTerm2|✳ Check today's app usage|2026-04-09T17:09:23.768571+00:00
iTerm2|✳ Check today's app usage|2026-04-09T17:09:12.596384+00:00
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ cd ~/.screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ ll
total 55496
drwxr-xr-x 8 lukas staff 256 9 Apr 19:53 .
drwx------+ 88 lukas staff 2816 9 Apr 20:09 ..
drwxr-xr-x 74 lukas staff 2368 9 Apr 20:10 data
-rw-r--r-- 1 lukas staff 15765504 9 Apr 20:08 db.sqlite
-rw-r--r-- 1 lukas staff 32768 9 Apr 19:53 db.sqlite-shm
-rw-r--r-- 1 lukas staff 11696712 9 Apr 20:11 db.sqlite-wal
drwxr-xr-x 8 lukas staff 256 9 Apr 19:53 pipes
-rw-r--r-- 1 lukas staff 25189 9 Apr 20:09 screenpipe.2026-04-09.0.log
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ cat > ~/.screenpipe/config.json << 'EOF'
{
"disable_audio": true
}
EOF
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ ll
total 55504
drwxr-xr-x 9 lukas staff 288 9 Apr 20:11 .
drwx------+ 88 lukas staff 2816 9 Apr 20:09 ..
-rw-r--r-- 1 lukas staff 28 9 Apr 20:11 config.json
drwxr-xr-x 74 lukas staff 2368 9 Apr 20:10 data
-rw-r--r-- 1 lukas staff 15765504 9 Apr 20:08 db.sqlite
-rw-r--r-- 1 lukas staff 32768 9 Apr 19:53 db.sqlite-shm
-rw-r--r-- 1 lukas staff 12195232 9 Apr 20:11 db.sqlite-wal
drwxr-xr-x 8 lukas staff 256 9 Apr 19:53 pipes
-rw-r--r-- 1 lukas staff 25189 9 Apr 20:09 screenpipe.2026-04-09.0.log
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ nano config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ nano config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ ls -la ~/.screenpipe/config.json
-rw-r--r-- 1 lukas staff 177 9 Apr 20:13 /Users/lukas/.screenpipe/config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ pkill -f screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ npx screenpipe@latest record --disable-audio
detected hardware tier: Mid
checking permissions...
screen recording: ok
accessibility: ok
2026-04-09T20:19:27.531337Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:19:28.280889Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:19:28.282336Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:19:28.282354Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:19:28.282617Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:19:28.282705Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:19:28.282660Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:19:28.283417Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:19:28.283684Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:19:28.284452Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:19:28.284646Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:19:28.284821Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:19:28.285013Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:19:28.285763Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:19:28.285784Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ true │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ disabled │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: [URL_WITH_CREDENTIALS] ~/.screenpipe $ pkill -f screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ pkill -f screenpipe
UW PICO 5.09 New Buffer
[ Read 12 lines ]
^G Get Help ^O WriteOut ^R Read File ^Y Prev Pg ^K Cut Text ^C Cur Pos
^X Exit ^J Justify ^W Where is ^V Next Pg ^U UnCut Text ^T To Spell
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ pkill -f screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ cat ~/.screenpipe/config.json
{
"disable_audio": true,
"ignored_windows": [
"1Password",
"Keychain Access",
"Bitwarden",
"System Preferences",
"System Settings",
"zoom.us"
]
}
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ sqlite3 ~/.screenpipe/db.sqlite "SELECT * FROM monitors;" 2>/dev/null || echo "no monitors table"
no monitors table
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ curl [URL_WITH_CREDENTIALS] ~/.screenpipe $ cat > ~/.screenpipe/config.json << 'EOF'
{
"disable_audio": true,
"ignored_windows": [
"1Password",
"Keychain Access",
"Bitwarden",
"System Preferences",
"System Settings",
"zoom.us"
],
"monitor_ids": ["monitor_1", "monitor_2"]
}
EOF
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ nano config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ curl [URL_WITH_CREDENTIALS] ~/.screenpipe $ cat > ~/.screenpipe/config.json << 'EOF'
{
"disable_audio": true,
"monitor_ids": ["Display 1_1440x900_0,0", "Display 2_3008x1253_-813,-1253"],
"ignored_windows": [
"1Password",
"Keychain Access",
"Bitwarden",
"System Preferences",
"System Settings",
"zoom.us"
]
}
EOF
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ pkill -f screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ npx screenpipe@latest record --disable-audio
detected hardware tier: Mid
checking permissions...
screen recording: ok
accessibility: ok
2026-04-09T20:25:00.748737Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:25:01.402463Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:25:01.404239Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:25:01.404260Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:25:01.404613Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:25:01.404686Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:25:01.404708Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:25:01.404722Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:25:01.405384Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:25:01.405652Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:25:01.406204Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:25:01.406386Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:25:01.406584Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:25:01.407568Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:25:01.407611Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ true │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
2026-04-09T20:25:01.409910Z INFO screenpipe_core::pipes: pipe scheduler started (generation 2)
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ disabled │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: https://github.com/screenpipe/screenpipe/releases
2026-04-09T20:25:01.410180Z INFO screenpipe: starting UI event capture
2026-04-09T20:25:01.413680Z WARN screenpipe: pi agent install failed: bun not found — install from https://bun.sh
2026-04-09T20:25:01.416008Z INFO screenpipe_engine::power::manager: initial power profile: Performance (on_ac=true, battery=Some(100))
2026-04-09T20:25:01.426030Z INFO screenpipe_engine::ui_recorder: Starting UI event capture
2026-04-09T20:25:01.442159Z INFO screenpipe_engine::calendar_speaker_id: speaker identification: started (user_name=<not set>)
2026-04-09T20:25:01.442149Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warming from DB (2026-04-08 17:25:01.442145 UTC to 2026-04-09 17:25:01.442145 UTC)
2026-04-09T20:25:01.442153Z INFO screenpipe_engine::ui_recorder: UI recording session started: 73715fed-9199-4fbb-acd0-0fb2b94e08d7
2026-04-09T20:25:01.442919Z INFO screenpipe_engine::meeting_detector: meeting v2: detection loop started (base_interval=5s, profiles=12)
2026-04-09T20:25:01.450915Z INFO screenpipe_engine::server: Server listening on [IP_ADDRESS]:3030
2026-04-09T20:25:01.455878Z INFO screenpipe_connect::mdns: mdns: advertising screenpipe on port 3030
2026-04-09T20:25:01.521290Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warmed with 242 frame entries, coverage from 2026-04-08 17:25:01.442145 UTC
2026-04-09T20:25:01.808367Z INFO screenpipe_engine::vision_manager::manager: Starting vision recording for monitor 1 (1440x900)
2026-04-09T20:25:01.808393Z INFO screenpipe_engine::vision_manager::manager: Starting event-driven capture for monitor 1 (device: monitor_1)
2026-04-09T20:25:01.808403Z INFO screenpipe_engine::vision_manager::manager: Skipping monitor 2 (Display 2_3008x1253_-813,-1253) — not in allowed list
2026-04-09T20:25:01.808411Z INFO screenpipe_engine::vision_manager::monitor_watcher: Starting monitor watcher (polling every 5 seconds)
2026-04-09T20:25:01.808511Z INFO screenpipe_engine::event_driven_capture: event-driven capture started for monitor 1 (device: monitor_1)
2026-04-09T20:25:02.610368Z INFO sck_rs::stream_manager: persistent SCK stream started for display 1 (1440x900, 2fps)
DOCKER
Close Tab
DEV (-zsh)
Close Tab
APP (-zsh)
Close Tab
✳ Check today's app usage (claude)
Close Tab
-zsh
Close Tab
screenpipe"
Close Tab
⌥⌘1
screenpipe"...
|
iTerm2
|
screenpipe"
|
NULL
|
|
UW PICO 5.09 UW PICO 5.09 New Buffer
[ Read 3 lines ]
^G Get Help ^O WriteOut ^R Read File ^Y Prev Pg ^K Cut Text ^C Cur Pos
^X Exit ^J Justify ^W Where is ^V Next Pg ^U UnCut Text ^T To Spell
Last login: Thu Apr 9 19:54:42 on ttys010
UW PICO 5.09 New Buffer
[ Read 11 lines ]
^G Get Help ^O WriteOut ^R Read File ^Y Prev Pg ^K Cut Text ^C Cur Pos
^X Exit ^J Justify ^W Where is ^V Next Pg ^U UnCut Text ^T To Spell
Poetry could not find a pyproject.toml file in /Users/lukas or its parents
Poetry could not find a pyproject.toml file in /Users/lukas or its parents
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ sqlite3 ~/.screenpipe/db.sqlite "SELECT app_name, window_name, timestamp FROM frames ORDER BY timestamp DESC LIMIT 10;"
iTerm2|-zsh|2026-04-09T17:09:56.277988+00:00
||2026-04-09T17:09:54.152728+00:00
iTerm2|✳ Check today's app usage|2026-04-09T17:09:52.277880+00:00
||2026-04-09T17:09:50.551186+00:00
||2026-04-09T17:09:48.080710+00:00
||2026-04-09T17:09:44.554361+00:00
||2026-04-09T17:09:42.248132+00:00
||2026-04-09T17:09:38.373309+00:00
iTerm2|✳ Check today's app usage|2026-04-09T17:09:23.768571+00:00
iTerm2|✳ Check today's app usage|2026-04-09T17:09:12.596384+00:00
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ cd ~/.screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ ll
total 55496
drwxr-xr-x 8 lukas staff 256 9 Apr 19:53 .
drwx------+ 88 lukas staff 2816 9 Apr 20:09 ..
drwxr-xr-x 74 lukas staff 2368 9 Apr 20:10 data
-rw-r--r-- 1 lukas staff 15765504 9 Apr 20:08 db.sqlite
-rw-r--r-- 1 lukas staff 32768 9 Apr 19:53 db.sqlite-shm
-rw-r--r-- 1 lukas staff 11696712 9 Apr 20:11 db.sqlite-wal
drwxr-xr-x 8 lukas staff 256 9 Apr 19:53 pipes
-rw-r--r-- 1 lukas staff 25189 9 Apr 20:09 screenpipe.2026-04-09.0.log
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ cat > ~/.screenpipe/config.json << 'EOF'
{
"disable_audio": true
}
EOF
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ ll
total 55504
drwxr-xr-x 9 lukas staff 288 9 Apr 20:11 .
drwx------+ 88 lukas staff 2816 9 Apr 20:09 ..
-rw-r--r-- 1 lukas staff 28 9 Apr 20:11 config.json
drwxr-xr-x 74 lukas staff 2368 9 Apr 20:10 data
-rw-r--r-- 1 lukas staff 15765504 9 Apr 20:08 db.sqlite
-rw-r--r-- 1 lukas staff 32768 9 Apr 19:53 db.sqlite-shm
-rw-r--r-- 1 lukas staff 12195232 9 Apr 20:11 db.sqlite-wal
drwxr-xr-x 8 lukas staff 256 9 Apr 19:53 pipes
-rw-r--r-- 1 lukas staff 25189 9 Apr 20:09 screenpipe.2026-04-09.0.log
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ nano config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ nano config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ ls -la ~/.screenpipe/config.json
-rw-r--r-- 1 lukas staff 177 9 Apr 20:13 /Users/lukas/.screenpipe/config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ pkill -f screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ npx screenpipe@latest record --disable-audio
detected hardware tier: Mid
checking permissions...
screen recording: ok
accessibility: ok
2026-04-09T20:19:27.531337Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:19:28.280889Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:19:28.282336Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:19:28.282354Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:19:28.282617Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:19:28.282705Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:19:28.282660Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:19:28.283417Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:19:28.283684Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:19:28.284452Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:19:28.284646Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:19:28.284821Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:19:28.285013Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:19:28.285763Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:19:28.285784Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ true │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ disabled │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: [URL_WITH_CREDENTIALS] ~/.screenpipe $ pkill -f screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ pkill -f screenpipe
UW PICO 5.09 New Buffer
[ Read 12 lines ]
^G Get Help ^O WriteOut ^R Read File ^Y Prev Pg ^K Cut Text ^C Cur Pos
^X Exit ^J Justify ^W Where is ^V Next Pg ^U UnCut Text ^T To Spell
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ pkill -f screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ cat ~/.screenpipe/config.json
{
"disable_audio": true,
"ignored_windows": [
"1Password",
"Keychain Access",
"Bitwarden",
"System Preferences",
"System Settings",
"zoom.us"
]
}
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ sqlite3 ~/.screenpipe/db.sqlite "SELECT * FROM monitors;" 2>/dev/null || echo "no monitors table"
no monitors table
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ curl [URL_WITH_CREDENTIALS] ~/.screenpipe $ cat > ~/.screenpipe/config.json << 'EOF'
{
"disable_audio": true,
"ignored_windows": [
"1Password",
"Keychain Access",
"Bitwarden",
"System Preferences",
"System Settings",
"zoom.us"
],
"monitor_ids": ["monitor_1", "monitor_2"]
}
EOF
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ nano config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ curl [URL_WITH_CREDENTIALS] ~/.screenpipe $ cat > ~/.screenpipe/config.json << 'EOF'
{
"disable_audio": true,
"monitor_ids": ["Display 1_1440x900_0,0", "Display 2_3008x1253_-813,-1253"],
"ignored_windows": [
"1Password",
"Keychain Access",
"Bitwarden",
"System Preferences",
"System Settings",
"zoom.us"
]
}
EOF
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ pkill -f screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ npx screenpipe@latest record --disable-audio
detected hardware tier: Mid
checking permissions...
screen recording: ok
accessibility: ok
2026-04-09T20:25:00.748737Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:25:01.402463Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:25:01.404239Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:25:01.404260Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:25:01.404613Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:25:01.404686Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:25:01.404708Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:25:01.404722Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:25:01.405384Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:25:01.405652Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:25:01.406204Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:25:01.406386Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:25:01.406584Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:25:01.407568Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:25:01.407611Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ true │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
2026-04-09T20:25:01.409910Z INFO screenpipe_core::pipes: pipe scheduler started (generation 2)
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ disabled │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: https://github.com/screenpipe/screenpipe/releases
2026-04-09T20:25:01.410180Z INFO screenpipe: starting UI event capture
2026-04-09T20:25:01.413680Z WARN screenpipe: pi agent install failed: bun not found — install from https://bun.sh
2026-04-09T20:25:01.416008Z INFO screenpipe_engine::power::manager: initial power profile: Performance (on_ac=true, battery=Some(100))
2026-04-09T20:25:01.426030Z INFO screenpipe_engine::ui_recorder: Starting UI event capture
2026-04-09T20:25:01.442159Z INFO screenpipe_engine::calendar_speaker_id: speaker identification: started (user_name=<not set>)
2026-04-09T20:25:01.442149Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warming from DB (2026-04-08 17:25:01.442145 UTC to 2026-04-09 17:25:01.442145 UTC)
2026-04-09T20:25:01.442153Z INFO screenpipe_engine::ui_recorder: UI recording session started: 73715fed-9199-4fbb-acd0-0fb2b94e08d7
2026-04-09T20:25:01.442919Z INFO screenpipe_engine::meeting_detector: meeting v2: detection loop started (base_interval=5s, profiles=12)
2026-04-09T20:25:01.450915Z INFO screenpipe_engine::server: Server listening on [IP_ADDRESS]:3030
2026-04-09T20:25:01.455878Z INFO screenpipe_connect::mdns: mdns: advertising screenpipe on port 3030
2026-04-09T20:25:01.521290Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warmed with 242 frame entries, coverage from 2026-04-08 17:25:01.442145 UTC
2026-04-09T20:25:01.808367Z INFO screenpipe_engine::vision_manager::manager: Starting vision recording for monitor 1 (1440x900)
2026-04-09T20:25:01.808393Z INFO screenpipe_engine::vision_manager::manager: Starting event-driven capture for monitor 1 (device: monitor_1)
2026-04-09T20:25:01.808403Z INFO screenpipe_engine::vision_manager::manager: Skipping monitor 2 (Display 2_3008x1253_-813,-1253) — not in allowed list
2026-04-09T20:25:01.808411Z INFO screenpipe_engine::vision_manager::monitor_watcher: Starting monitor watcher (polling every 5 seconds)
2026-04-09T20:25:01.808511Z INFO screenpipe_engine::event_driven_capture: event-driven capture started for monitor 1 (device: monitor_1)
2026-04-09T20:25:02.610368Z INFO sck_rs::stream_manager: persistent SCK stream started for display 1 (1440x900, 2fps)
2026-04-09T20:25:06.039725Z INFO screenpipe_engine::event_driven_capture: startup capture for monitor 1: frame_id=243, dur=50ms
2026-04-09T20:25:06.445466Z INFO screenpipe_engine::retention: retention: initialized with 14d retention
2026-04-09T20:25:06.445759Z INFO screenpipe: local retention auto-enabled (14 days)
DOCKER
Close Tab
DEV (-zsh)
Close Tab
APP (-zsh)
Close Tab
✳ Check today's app usage (claude)
Close Tab
-zsh
Close Tab
screenpipe"
Close Tab
⌥⌘1
screenpipe"...
|
iTerm2
|
screenpipe"
|
NULL
|
|
UW PICO 5.09 UW PICO 5.09 New Buffer
[ Read 3 lines ]
^G Get Help ^O WriteOut ^R Read File ^Y Prev Pg ^K Cut Text ^C Cur Pos
^X Exit ^J Justify ^W Where is ^V Next Pg ^U UnCut Text ^T To Spell
Last login: Thu Apr 9 19:54:42 on ttys010
UW PICO 5.09 New Buffer
[ Read 11 lines ]
^G Get Help ^O WriteOut ^R Read File ^Y Prev Pg ^K Cut Text ^C Cur Pos
^X Exit ^J Justify ^W Where is ^V Next Pg ^U UnCut Text ^T To Spell
Poetry could not find a pyproject.toml file in /Users/lukas or its parents
Poetry could not find a pyproject.toml file in /Users/lukas or its parents
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ sqlite3 ~/.screenpipe/db.sqlite "SELECT app_name, window_name, timestamp FROM frames ORDER BY timestamp DESC LIMIT 10;"
iTerm2|-zsh|2026-04-09T17:09:56.277988+00:00
||2026-04-09T17:09:54.152728+00:00
iTerm2|✳ Check today's app usage|2026-04-09T17:09:52.277880+00:00
||2026-04-09T17:09:50.551186+00:00
||2026-04-09T17:09:48.080710+00:00
||2026-04-09T17:09:44.554361+00:00
||2026-04-09T17:09:42.248132+00:00
||2026-04-09T17:09:38.373309+00:00
iTerm2|✳ Check today's app usage|2026-04-09T17:09:23.768571+00:00
iTerm2|✳ Check today's app usage|2026-04-09T17:09:12.596384+00:00
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ cd ~/.screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ ll
total 55496
drwxr-xr-x 8 lukas staff 256 9 Apr 19:53 .
drwx------+ 88 lukas staff 2816 9 Apr 20:09 ..
drwxr-xr-x 74 lukas staff 2368 9 Apr 20:10 data
-rw-r--r-- 1 lukas staff 15765504 9 Apr 20:08 db.sqlite
-rw-r--r-- 1 lukas staff 32768 9 Apr 19:53 db.sqlite-shm
-rw-r--r-- 1 lukas staff 11696712 9 Apr 20:11 db.sqlite-wal
drwxr-xr-x 8 lukas staff 256 9 Apr 19:53 pipes
-rw-r--r-- 1 lukas staff 25189 9 Apr 20:09 screenpipe.2026-04-09.0.log
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ cat > ~/.screenpipe/config.json << 'EOF'
{
"disable_audio": true
}
EOF
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ ll
total 55504
drwxr-xr-x 9 lukas staff 288 9 Apr 20:11 .
drwx------+ 88 lukas staff 2816 9 Apr 20:09 ..
-rw-r--r-- 1 lukas staff 28 9 Apr 20:11 config.json
drwxr-xr-x 74 lukas staff 2368 9 Apr 20:10 data
-rw-r--r-- 1 lukas staff 15765504 9 Apr 20:08 db.sqlite
-rw-r--r-- 1 lukas staff 32768 9 Apr 19:53 db.sqlite-shm
-rw-r--r-- 1 lukas staff 12195232 9 Apr 20:11 db.sqlite-wal
drwxr-xr-x 8 lukas staff 256 9 Apr 19:53 pipes
-rw-r--r-- 1 lukas staff 25189 9 Apr 20:09 screenpipe.2026-04-09.0.log
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ nano config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ nano config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ ls -la ~/.screenpipe/config.json
-rw-r--r-- 1 lukas staff 177 9 Apr 20:13 /Users/lukas/.screenpipe/config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ pkill -f screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ npx screenpipe@latest record --disable-audio
detected hardware tier: Mid
checking permissions...
screen recording: ok
accessibility: ok
2026-04-09T20:19:27.531337Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:19:28.280889Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:19:28.282336Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:19:28.282354Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:19:28.282617Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:19:28.282705Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:19:28.282660Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:19:28.283417Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:19:28.283684Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:19:28.284452Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:19:28.284646Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:19:28.284821Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:19:28.285013Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:19:28.285763Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:19:28.285784Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ true │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ disabled │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: [URL_WITH_CREDENTIALS] ~/.screenpipe $ pkill -f screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ pkill -f screenpipe
UW PICO 5.09 New Buffer
[ Read 12 lines ]
^G Get Help ^O WriteOut ^R Read File ^Y Prev Pg ^K Cut Text ^C Cur Pos
^X Exit ^J Justify ^W Where is ^V Next Pg ^U UnCut Text ^T To Spell
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ pkill -f screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ cat ~/.screenpipe/config.json
{
"disable_audio": true,
"ignored_windows": [
"1Password",
"Keychain Access",
"Bitwarden",
"System Preferences",
"System Settings",
"zoom.us"
]
}
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ sqlite3 ~/.screenpipe/db.sqlite "SELECT * FROM monitors;" 2>/dev/null || echo "no monitors table"
no monitors table
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ curl [URL_WITH_CREDENTIALS] ~/.screenpipe $ cat > ~/.screenpipe/config.json << 'EOF'
{
"disable_audio": true,
"ignored_windows": [
"1Password",
"Keychain Access",
"Bitwarden",
"System Preferences",
"System Settings",
"zoom.us"
],
"monitor_ids": ["monitor_1", "monitor_2"]
}
EOF
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ nano config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ curl [URL_WITH_CREDENTIALS] ~/.screenpipe $ cat > ~/.screenpipe/config.json << 'EOF'
{
"disable_audio": true,
"monitor_ids": ["Display 1_1440x900_0,0", "Display 2_3008x1253_-813,-1253"],
"ignored_windows": [
"1Password",
"Keychain Access",
"Bitwarden",
"System Preferences",
"System Settings",
"zoom.us"
]
}
EOF
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ pkill -f screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ npx screenpipe@latest record --disable-audio
detected hardware tier: Mid
checking permissions...
screen recording: ok
accessibility: ok
2026-04-09T20:25:00.748737Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:25:01.402463Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:25:01.404239Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:25:01.404260Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:25:01.404613Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:25:01.404686Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:25:01.404708Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:25:01.404722Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:25:01.405384Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:25:01.405652Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:25:01.406204Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:25:01.406386Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:25:01.406584Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:25:01.407568Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:25:01.407611Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ true │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
2026-04-09T20:25:01.409910Z INFO screenpipe_core::pipes: pipe scheduler started (generation 2)
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ disabled │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: https://github.com/screenpipe/screenpipe/releases
2026-04-09T20:25:01.410180Z INFO screenpipe: starting UI event capture
2026-04-09T20:25:01.413680Z WARN screenpipe: pi agent install failed: bun not found — install from https://bun.sh
2026-04-09T20:25:01.416008Z INFO screenpipe_engine::power::manager: initial power profile: Performance (on_ac=true, battery=Some(100))
2026-04-09T20:25:01.426030Z INFO screenpipe_engine::ui_recorder: Starting UI event capture
2026-04-09T20:25:01.442159Z INFO screenpipe_engine::calendar_speaker_id: speaker identification: started (user_name=<not set>)
2026-04-09T20:25:01.442149Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warming from DB (2026-04-08 17:25:01.442145 UTC to 2026-04-09 17:25:01.442145 UTC)
2026-04-09T20:25:01.442153Z INFO screenpipe_engine::ui_recorder: UI recording session started: 73715fed-9199-4fbb-acd0-0fb2b94e08d7
2026-04-09T20:25:01.442919Z INFO screenpipe_engine::meeting_detector: meeting v2: detection loop started (base_interval=5s, profiles=12)
2026-04-09T20:25:01.450915Z INFO screenpipe_engine::server: Server listening on [IP_ADDRESS]:3030
2026-04-09T20:25:01.455878Z INFO screenpipe_connect::mdns: mdns: advertising screenpipe on port 3030
2026-04-09T20:25:01.521290Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warmed with 242 frame entries, coverage from 2026-04-08 17:25:01.442145 UTC
2026-04-09T20:25:01.808367Z INFO screenpipe_engine::vision_manager::manager: Starting vision recording for monitor 1 (1440x900)
2026-04-09T20:25:01.808393Z INFO screenpipe_engine::vision_manager::manager: Starting event-driven capture for monitor 1 (device: monitor_1)
2026-04-09T20:25:01.808403Z INFO screenpipe_engine::vision_manager::manager: Skipping monitor 2 (Display 2_3008x1253_-813,-1253) — not in allowed list
2026-04-09T20:25:01.808411Z INFO screenpipe_engine::vision_manager::monitor_watcher: Starting monitor watcher (polling every 5 seconds)
2026-04-09T20:25:01.808511Z INFO screenpipe_engine::event_driven_capture: event-driven capture started for monitor 1 (device: monitor_1)
2026-04-09T20:25:02.610368Z INFO sck_rs::stream_manager: persistent SCK stream started for display 1 (1440x900, 2fps)
2026-04-09T20:25:06.039725Z INFO screenpipe_engine::event_driven_capture: startup capture for monitor 1: frame_id=243, dur=50ms
2026-04-09T20:25:06.445466Z INFO screenpipe_engine::retention: retention: initialized with 14d retention
2026-04-09T20:25:06.445759Z INFO screenpipe: local retention auto-enabled (14 days)
Starting vision recording for monitor 1 (1440x900)
search
cancel
go left
go right
DOCKER
Close Tab
DEV (-zsh)
Close Tab
APP (-zsh)
Close Tab
✳ Check today's app usage (claude)
Close Tab
-zsh
Close Tab
screenpipe"
Close Tab
⌥⌘1
screenpipe"...
|
iTerm2
|
screenpipe"
|
NULL
|
|
UW PICO 5.09 UW PICO 5.09 New Buffer
[ Read 3 lines ]
^G Get Help ^O WriteOut ^R Read File ^Y Prev Pg ^K Cut Text ^C Cur Pos
^X Exit ^J Justify ^W Where is ^V Next Pg ^U UnCut Text ^T To Spell
Last login: Thu Apr 9 19:54:42 on ttys010
UW PICO 5.09 New Buffer
[ Read 11 lines ]
^G Get Help ^O WriteOut ^R Read File ^Y Prev Pg ^K Cut Text ^C Cur Pos
^X Exit ^J Justify ^W Where is ^V Next Pg ^U UnCut Text ^T To Spell
Poetry could not find a pyproject.toml file in /Users/lukas or its parents
Poetry could not find a pyproject.toml file in /Users/lukas or its parents
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ sqlite3 ~/.screenpipe/db.sqlite "SELECT app_name, window_name, timestamp FROM frames ORDER BY timestamp DESC LIMIT 10;"
iTerm2|-zsh|2026-04-09T17:09:56.277988+00:00
||2026-04-09T17:09:54.152728+00:00
iTerm2|✳ Check today's app usage|2026-04-09T17:09:52.277880+00:00
||2026-04-09T17:09:50.551186+00:00
||2026-04-09T17:09:48.080710+00:00
||2026-04-09T17:09:44.554361+00:00
||2026-04-09T17:09:42.248132+00:00
||2026-04-09T17:09:38.373309+00:00
iTerm2|✳ Check today's app usage|2026-04-09T17:09:23.768571+00:00
iTerm2|✳ Check today's app usage|2026-04-09T17:09:12.596384+00:00
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~ $ cd ~/.screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ ll
total 55496
drwxr-xr-x 8 lukas staff 256 9 Apr 19:53 .
drwx------+ 88 lukas staff 2816 9 Apr 20:09 ..
drwxr-xr-x 74 lukas staff 2368 9 Apr 20:10 data
-rw-r--r-- 1 lukas staff 15765504 9 Apr 20:08 db.sqlite
-rw-r--r-- 1 lukas staff 32768 9 Apr 19:53 db.sqlite-shm
-rw-r--r-- 1 lukas staff 11696712 9 Apr 20:11 db.sqlite-wal
drwxr-xr-x 8 lukas staff 256 9 Apr 19:53 pipes
-rw-r--r-- 1 lukas staff 25189 9 Apr 20:09 screenpipe.2026-04-09.0.log
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ cat > ~/.screenpipe/config.json << 'EOF'
{
"disable_audio": true
}
EOF
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ ll
total 55504
drwxr-xr-x 9 lukas staff 288 9 Apr 20:11 .
drwx------+ 88 lukas staff 2816 9 Apr 20:09 ..
-rw-r--r-- 1 lukas staff 28 9 Apr 20:11 config.json
drwxr-xr-x 74 lukas staff 2368 9 Apr 20:10 data
-rw-r--r-- 1 lukas staff 15765504 9 Apr 20:08 db.sqlite
-rw-r--r-- 1 lukas staff 32768 9 Apr 19:53 db.sqlite-shm
-rw-r--r-- 1 lukas staff 12195232 9 Apr 20:11 db.sqlite-wal
drwxr-xr-x 8 lukas staff 256 9 Apr 19:53 pipes
-rw-r--r-- 1 lukas staff 25189 9 Apr 20:09 screenpipe.2026-04-09.0.log
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ nano config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ nano config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ ls -la ~/.screenpipe/config.json
-rw-r--r-- 1 lukas staff 177 9 Apr 20:13 /Users/lukas/.screenpipe/config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ pkill -f screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ npx screenpipe@latest record --disable-audio
detected hardware tier: Mid
checking permissions...
screen recording: ok
accessibility: ok
2026-04-09T20:19:27.531337Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:19:28.280889Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:19:28.282336Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:19:28.282354Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:19:28.282617Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:19:28.282705Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:19:28.282660Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:19:28.283417Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:19:28.283684Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:19:28.284452Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:19:28.284646Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:19:28.284821Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:19:28.285013Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:19:28.285763Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:19:28.285784Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ true │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ disabled │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: [URL_WITH_CREDENTIALS] ~/.screenpipe $ pkill -f screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ pkill -f screenpipe
UW PICO 5.09 New Buffer
[ Read 12 lines ]
^G Get Help ^O WriteOut ^R Read File ^Y Prev Pg ^K Cut Text ^C Cur Pos
^X Exit ^J Justify ^W Where is ^V Next Pg ^U UnCut Text ^T To Spell
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ pkill -f screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ cat ~/.screenpipe/config.json
{
"disable_audio": true,
"ignored_windows": [
"1Password",
"Keychain Access",
"Bitwarden",
"System Preferences",
"System Settings",
"zoom.us"
]
}
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ sqlite3 ~/.screenpipe/db.sqlite "SELECT * FROM monitors;" 2>/dev/null || echo "no monitors table"
no monitors table
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ curl [URL_WITH_CREDENTIALS] ~/.screenpipe $ cat > ~/.screenpipe/config.json << 'EOF'
{
"disable_audio": true,
"ignored_windows": [
"1Password",
"Keychain Access",
"Bitwarden",
"System Preferences",
"System Settings",
"zoom.us"
],
"monitor_ids": ["monitor_1", "monitor_2"]
}
EOF
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ nano config.json
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ curl [URL_WITH_CREDENTIALS] ~/.screenpipe $ cat > ~/.screenpipe/config.json << 'EOF'
{
"disable_audio": true,
"monitor_ids": ["Display 1_1440x900_0,0", "Display 2_3008x1253_-813,-1253"],
"ignored_windows": [
"1Password",
"Keychain Access",
"Bitwarden",
"System Preferences",
"System Settings",
"zoom.us"
]
}
EOF
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ pkill -f screenpipe
lukas@Lukas-Kovaliks-MacBook-Pro-Jiminny ~/.screenpipe $ npx screenpipe@latest record --disable-audio
detected hardware tier: Mid
checking permissions...
screen recording: ok
accessibility: ok
2026-04-09T20:25:00.748737Z INFO screenpipe_screen::monitor::macos_version: Detected macOS version: 14.6
2026-04-09T20:25:01.402463Z INFO screenpipe_engine::sleep_monitor: Starting macOS sleep/wake monitor
2026-04-09T20:25:01.404239Z INFO screenpipe: meeting detector enabled — independent of transcription mode
2026-04-09T20:25:01.404260Z INFO screenpipe_engine::sleep_monitor: Screen lock/unlock observers registered (CFNotificationCenter)
2026-04-09T20:25:01.404613Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction worker started (min_age=600s, poll=300s)
2026-04-09T20:25:01.404686Z INFO screenpipe_engine::sleep_monitor: Display reconfiguration watcher registered (CGDisplayRegisterReconfigurationCallback)
2026-04-09T20:25:01.404708Z INFO screenpipe_engine::vision_manager::manager: Starting VisionManager
2026-04-09T20:25:01.404722Z INFO screenpipe_engine::power::manager: power manager started (poll interval: 10s)
2026-04-09T20:25:01.405384Z INFO screenpipe_core::pipes: loaded pipe: day-recap
2026-04-09T20:25:01.405652Z INFO screenpipe_core::pipes: loaded pipe: standup-update
2026-04-09T20:25:01.406204Z INFO screenpipe_core::pipes: loaded pipe: ai-habits
2026-04-09T20:25:01.406386Z INFO screenpipe_core::pipes: loaded pipe: time-breakdown
2026-04-09T20:25:01.406584Z INFO screenpipe_core::pipes: loaded pipe: video-export
2026-04-09T20:25:01.407568Z INFO screenpipe_core::pipes: loaded pipe: meeting-summary
2026-04-09T20:25:01.407611Z INFO screenpipe_core::pipes: loaded 6 pipes from "/Users/lukas/.screenpipe/pipes"
_
__________________ ___ ____ ____ (_____ ___
/ ___/ ___/ ___/ _ \/ _ \/ __ \ / __ \/ / __ \/ _ \
(__ / /__/ / / __/ __/ / / / / /_/ / / /_/ / __/
/____/\___/_/ \___/\___/_/ /_/ / .___/_/ .___/\___/
/_/ /_/
power AI by everything you've seen, said or heard
open source | runs locally | developer friendly
┌────────────────────────┬────────────────────────────────────┐
│ setting │ value │
├────────────────────────┼────────────────────────────────────┤
│ audio chunk duration │ 30 seconds │
│ port │ 3030 │
│ audio disabled │ true │
│ vision disabled │ false │
│ pause on DRM content │ false │
│ audio engine │ Parakeet │
│ vad engine │ Silero │
│ data directory │ /Users/lukas/.screenpipe │
│ debug mode │ false │
│ telemetry │ true │
│ use pii removal │ true │
│ use all monitors │ true │
│ ignored windows │ [] │
│ included windows │ [] │
│ cloud sync │ disabled │
│ auto-destruct pid │ 0 │
│ deepgram key │ not set │
├────────────────────────┼────────────────────────────────────┤
│ languages │ │
│ │ all languages │
├────────────────────────┼────────────────────────────────────┤
│ monitors │ │
2026-04-09T20:25:01.409910Z INFO screenpipe_core::pipes: pipe scheduler started (generation 2)
│ │ no monitors available │
├────────────────────────┼────────────────────────────────────┤
│ audio devices │ │
│ │ disabled │
└────────────────────────┴────────────────────────────────────┘
you are using local processing. all your data stays on your computer.
warning: telemetry is enabled. only error-level data will be sent.
to disable, use the --disable-telemetry flag.
check latest changes here: https://github.com/screenpipe/screenpipe/releases
2026-04-09T20:25:01.410180Z INFO screenpipe: starting UI event capture
2026-04-09T20:25:01.413680Z WARN screenpipe: pi agent install failed: bun not found — install from https://bun.sh
2026-04-09T20:25:01.416008Z INFO screenpipe_engine::power::manager: initial power profile: Performance (on_ac=true, battery=Some(100))
2026-04-09T20:25:01.426030Z INFO screenpipe_engine::ui_recorder: Starting UI event capture
2026-04-09T20:25:01.442159Z INFO screenpipe_engine::calendar_speaker_id: speaker identification: started (user_name=<not set>)
2026-04-09T20:25:01.442149Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warming from DB (2026-04-08 17:25:01.442145 UTC to 2026-04-09 17:25:01.442145 UTC)
2026-04-09T20:25:01.442153Z INFO screenpipe_engine::ui_recorder: UI recording session started: 73715fed-9199-4fbb-acd0-0fb2b94e08d7
2026-04-09T20:25:01.442919Z INFO screenpipe_engine::meeting_detector: meeting v2: detection loop started (base_interval=5s, profiles=12)
2026-04-09T20:25:01.450915Z INFO screenpipe_engine::server: Server listening on [IP_ADDRESS]:3030
2026-04-09T20:25:01.455878Z INFO screenpipe_connect::mdns: mdns: advertising screenpipe on port 3030
2026-04-09T20:25:01.521290Z INFO screenpipe_engine::hot_frame_cache: hot_frame_cache: warmed with 242 frame entries, coverage from 2026-04-08 17:25:01.442145 UTC
2026-04-09T20:25:01.808367Z INFO screenpipe_engine::vision_manager::manager: Starting vision recording for monitor 1 (1440x900)
2026-04-09T20:25:01.808393Z INFO screenpipe_engine::vision_manager::manager: Starting event-driven capture for monitor 1 (device: monitor_1)
2026-04-09T20:25:01.808403Z INFO screenpipe_engine::vision_manager::manager: Skipping monitor 2 (Display 2_3008x1253_-813,-1253) — not in allowed list
2026-04-09T20:25:01.808411Z INFO screenpipe_engine::vision_manager::monitor_watcher: Starting monitor watcher (polling every 5 seconds)
2026-04-09T20:25:01.808511Z INFO screenpipe_engine::event_driven_capture: event-driven capture started for monitor 1 (device: monitor_1)
2026-04-09T20:25:02.610368Z INFO sck_rs::stream_manager: persistent SCK stream started for display 1 (1440x900, 2fps)
2026-04-09T20:25:06.039725Z INFO screenpipe_engine::event_driven_capture: startup capture for monitor 1: frame_id=243, dur=50ms
2026-04-09T20:25:06.445466Z INFO screenpipe_engine::retention: retention: initialized with 14d retention
2026-04-09T20:25:06.445759Z INFO screenpipe: local retention auto-enabled (14 days)
2026-04-09T20:25:27.448962Z INFO screenpipe_engine::event_driven_capture: content dedup: skipping capture for monitor 1 (hash=5187306990002451762, trigger=visual_change)
2026-04-09T20:26:01.408581Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: found 44 eligible frames
2026-04-09T20:26:05.200395Z INFO screenpipe_engine::snapshot_compaction: snapshot compaction: 43 frames, 15.1MB → 3.0MB (5.0x), 43 JPEGs deleted
Starting vision recording for monitor 1 (1440x900)
search
cancel
go left
go right
DOCKER
Close Tab
DEV (-zsh)
Close Tab
APP (-zsh)
Close Tab
✳ Check today's app usage (claude)
Close Tab
-zsh
Close Tab
screenpipe"
Close Tab
⌥⌘1
screenpipe"...
|
iTerm2
|
screenpipe"
|
NULL
|