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

..

conftest.py

test: lock speechbrain lengths and identity camera filter

2026-08-28 05:15:49 +00:00

test_audio_cleanup.py

perf: reuse GCC-PHAT TDOA; skip 16 kHz AudioNormalizer

2026-08-28 06:19:02 +00:00

test_config.py

perf: cap torch threads per camera worker

2026-08-28 06:17:33 +00:00

test_discovery.py

test: lock speechbrain lengths and identity camera filter

2026-08-28 05:15:49 +00:00