root
736eccb932
feat: known-good CPU audio + 1080p VAAPI + KMS display wall
...
Working snapshot before the OpenVINO MetricGAN GPU spike:
thread-capped SpeechBrain DelaySum+MetricGAN on CPU, hop 200 ms,
video hold for A/V sync, Arc VAAPI H.264, headless kmssink wall.
32 tests passing. Do not treat this as a GPU audio path.
2026-08-28 07:14:21 +00:00
root
11776adf52
chore: Arc B580 offload spike script
...
Read-only probe: CUDA unusable, OpenCL sees the B580, OpenVINO not
installed. Per-worker GPU still out of scope.
2026-08-28 06:26:02 +00:00
root
f83f0f0683
docs: latency budget, thread cap, and Arc B580 CUDA mismatch
...
Also extend bench_enhance.py with --stereo --hop for isolated RTF.
2026-08-28 06:21:02 +00:00
root
36c4bbc956
perf: ALSA ffmpeg low-delay flags
...
nobuffer, low_delay, and a small thread_queue_size on the camera
mic capture process.
2026-08-28 06:19:44 +00:00
root
b02138669a
perf: reuse GCC-PHAT TDOA; skip 16 kHz AudioNormalizer
...
Delay-and-Sum still uses the tutorial STFT/cov/GCC-PHAT/DelaySum/ISTFT
path. TDOA is refreshed every BEAMFORM_TDOA_EVERY hops (default 5).
2026-08-28 06:19:02 +00:00
root
23c2e9c4c0
perf: cap torch threads per camera worker
...
Default TORCH_NUM_THREADS=1 so 14-20 workers do not each spawn a
28-thread BLAS pool.
2026-08-28 06:17:33 +00:00
root
950128deec
feat: DelaySum beamform plus hop process latency
...
Keep C920 stereo through SpeechBrain GCC-PHAT Delay-and-Sum, then
MetricGAN, and publish mono to LiveKit. Record last_dt_ms and log
process_ms/capture_ms every 50 hops for the latency plan.
2026-08-28 06:16:25 +00:00
Hermes
abeea6256c
ops: appliance boot — wait for cameras/LiveKit, power button powers off
2026-08-28 05:45:11 +00:00
Hermes
b5a2fe6440
ops: persist livekit-server and camera fleet as systemd 24/7 units
2026-08-28 05:41:18 +00:00
Hermes
30b128bc9e
docs: serial identity, enhance hop/window, optional systemd unit
2026-08-28 05:21:36 +00:00
Hermes
106790cab8
test: smoke worker with speechbrain force on local livekit
2026-08-28 05:19:05 +00:00
Hermes
ad3b43eccd
fix: bench script sys.path so audio_cleanup imports
2026-08-28 05:17:32 +00:00
Hermes
d8452c0ffe
feat: overlap-window speechbrain enhance, hop 200ms / window 1s
2026-08-28 05:17:13 +00:00
Hermes
628307ff12
test: lock speechbrain lengths and identity camera filter
...
Filter --cameras 1,5 by cam-NN identity instead of discovery index.
2026-08-28 05:15:49 +00:00
Hermes
dc46cbef04
chore: snapshot livekit-cameras after serial pin + speechbrain lengths fix
2026-08-28 05:15:00 +00:00
root
Hermes