Commit Graph

5 Commits

Author SHA1 Message Date
Alex
22aae40d17 Add design doc for YAMNet sound identification on Coral Edge TPU
Covers model choice, architecture, category mapping, API endpoints,
and integration with existing headmic audio pipeline.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-01 20:04:31 -06:00
Alex Kazaiev
c6e18738ae Use device name instead of card number for ALSA
Card numbers can shift based on USB enumeration order at boot.
Using 'plughw:ArrayUAC10,0' instead of 'plughw:2,0' ensures
the ReSpeaker is found regardless of when it connects.

Fixed by Vixy after power loss shuffled card order 🦊
2026-01-21 12:20:39 -06:00
Alex Kazaiev
c53556fe97 Fix ReSpeaker device index: card 3 → card 2
USB device enumeration changed after GPIO rewiring for I2S audio.
TODO: Consider udev rule for stable device naming.
2026-01-17 16:20:15 -06:00
5ed2c6aee7 Fix: Use arecord for shared audio stream
- Replaced PyAudio with direct ALSA (arecord subprocess)
- Single audio stream feeds both Porcupine and recording buffer
- Fixes device unavailable error when recording after wake word
- Simplified architecture
2026-01-17 11:17:17 -06:00
be7e26b6e7 Initial commit: HeadMic service - Vixy's Ears 🦊👂
Wake word detection (Hey Vivi) + voice recording + EarTail transcription
Built by Vixy on Day 77
2026-01-17 10:58:51 -06:00