Unrecognized speakers now get stable IDs like "unknown_a7f3" instead
of None. Uses online clustering of Resemblyzer embeddings:
- Matches against tracked anonymous speakers (cosine > 0.70)
- Updates running average embedding on re-identification
- Creates new ID from SHA-256 hash of quantized embedding
- Expires after 1 hour of silence, max 10 tracked simultaneously
New API: POST /speakers/promote?anon_id=unknown_a7f3&name=Alex
Promotes an anonymous speaker to enrolled using their averaged embedding.
Flow: unknown person speaks → "unknown_a7f3" → you ask "who's that?" →
promote to "Bob" → now recognized by name going forward.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Cross-correlates left/right ear audio frames (512 samples, ~32ms window)
to find the sub-millisecond delay between arrays. Converts delay to
bearing angle using speed of sound and array separation.
At 16kHz with 175mm separation, resolution is ~1 sample = 62.5μs = ~7°.
Not lab-grade, but adds a third independent angle estimate alongside
DoA and ILD. Works with current 2-channel firmware — no raw mics needed.
New fields in /doa spatial response:
itd_angle: bearing from cross-correlation (degrees)
itd_delay_us: raw time delay (microseconds, positive = source on right)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
audio_stream.py: Added focus_side property. When set, the stream
yields from the focused side regardless of energy (attention lock).
When None, falls back to energy-based auto selection.
multi_speaker.py: When beams lock onto 2 speakers, sets audio focus
to the target speaker's side. Auto-switches target when the current
target goes silent and the other starts talking. Manual focus via API.
headmic.py: New endpoint POST /speakers/focus?speaker=0|1 to manually
switch attention. /speakers/tracked now shows is_target, target_speaker,
and audio_focus fields.
The cocktail party effect: when 2 people are talking, the audio feed
to Porcupine/VAD/transcription comes from the target speaker's direction,
suppressing the other. XVF3800 beam gating silences the non-speaking beam,
and audio_stream focus locks the ear facing the target.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
multi_speaker.py: Tracks up to 2 speakers simultaneously. When 2 distinct
DoA angles are detected (30°+ apart) for >1s, locks the XVF3800's fixed
beams onto each speaker. Releases back to auto mode when only 1 speaker
remains (3s timeout). Manages beam gating so only the speaking beam is active.
xvf3800.py: Added beam steering commands — enable_fixed_beams(),
set_beam_azimuths(), enable_beam_gating(), read_all_beams().
Manager gets steer_beams() and release_beams() convenience methods.
headmic.py: Wire multi-speaker tracker into DoA loop. New endpoint:
GET /speakers/tracked — current speaker positions, beam mode, lock state.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
10Hz DoA polling + 10Hz gaze HTTP pushes was creating too much
GIL pressure, starving uvicorn's async event loop. Reduced to
5Hz polling and max 2 gaze pushes/sec with 5px min delta.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Synchronous urllib.urlopen at 10Hz was starving uvicorn's event loop
via GIL contention. Now each push runs in its own daemon thread, and
skips if the previous push is still in flight (natural rate limiting).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
httpx.post creates a new connection per call at 10Hz, causing connection
pile-up that eventually blocks the event loop. urllib is lightweight and
stateless — no connection pooling overhead.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
spatial_scene.py: Builds a persistent map of where each sound category
usually comes from (30° angle bins, circular mean). Detects anomalies
when a sound appears from an unusual direction (90°+ deviation).
Scene map persists to ~/.vixy/scene_map.json across restarts.
headmic.py: Feed classified sounds + spatial position into scene tracker.
New endpoints:
/scene — learned scene summary + last anomaly
/scene/events — recent events with what+where+when
/scene/heatmap — per-category angular distribution (for visualization)
Example: after running for a day, /scene might show:
{"speech": {"usual_angle": 15.0, "observations": 847},
"music": {"usual_angle": 270.0, "observations": 312}}
And if speech comes from 270° (where music usually is): spatial anomaly.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Computes Interaural Level Difference (dB) from left/right ear energy.
Fuses with triangulated distance (70/30 weight) for more robust estimate.
Classifies into proximity zones: intimate (<0.5m), conversational (0.5-2m),
across_room (2-5m), far (>5m).
ILD→distance mapping is empirical and should be calibrated per install.
Gaze vertical component now responds to proximity (closer = eyes down).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
binaural_recorder.py: Records left/right ear streams as stereo WAV
in rolling 5-minute segments. Training data for spatial audio models.
Enabled via BINAURAL_RECORD=1 env var.
spatial.py: Tune smoothing — alpha 0.3→0.4 (snappier response),
idle return speed 0.05→0.03 (gentler drift), timeout 2s→1.5s.
headmic.py: Wire binaural recorder into audio loop, add /recording
endpoint for stats, feed both ear streams (not just best beam).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
startup() assigned spatial_tracker as a local variable instead of
updating the module-level global. doa_track_loop saw None.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
spatial.py: Triangulates sound source position from two DoA angles using
ray intersection. Exponential smoothing prevents jitter. Gaze drifts back
to center after 2s of silence. Converts position (mm) to gaze (0-255).
headmic.py: Replaces simple doa_poll_loop with doa_track_loop that runs
the spatial tracker and pushes gaze to the eye service when the position
changes. Rate-limited to 10 pushes/sec with minimum delta threshold.
/doa endpoint now returns triangulated position + gaze coordinates.
Array separation (175mm) stored in config, overridable.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Probes the Edge TPU in a subprocess before loading — catches segfaults
(libedgetpu ABI mismatch on Debian Trixie/Python 3.13) and falls back
to CPU automatically. No more service crashes on Coral incompatibility.
When the runtime is eventually fixed, Edge TPU will be used automatically
with no config changes needed.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
libedgetpu on Pi 5 segfaults with the compiled model.
CPU fallback works fine (~50-100ms at 0.5s intervals).
Set USE_EDGETPU=1 in headmic.service to enable once runtime is fixed.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Without this fix, listener_loop exits early on Porcupine init failure,
which starves the sound classifier ring buffer. Now the audio loop
continues for YAMNet classification even without wake word detection.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Prefer yamnet_edgetpu.tflite when available, fall back to CPU model.
~50-100ms → ~2-3ms inference per classification.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Adds voice-based speaker ID triggered by YAMNet speech detection.
New speaker_id.py module with SQLite-backed voice enrollment and
cosine similarity matching. Endpoints: POST /speakers/enroll,
POST /speakers/enroll-from-mic, GET /speakers, DELETE /speakers/{name}.
Orange LED animation during enrollment.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
New sound_id.py module with SoundClassifier class that runs YAMNet
(521 audio event categories) on CPU TFLite. Classifies audio every
0.5s from a ring buffer fed by the existing audio stream.
Categories: speech, alert, music, animal, household, environment, silence.
Smoothing via 20-sample history window for stable dominant category.
New endpoints: GET /sounds, GET /sounds/history
Updated: /health (sound_classification_enabled), /status (audio_scene)
Graceful degradation if model files not present.
Model download (not tracked in git):
curl -sL 'https://tfhub.dev/google/lite-model/yamnet/classification/tflite/1?lite-format=tflite' -o models/yamnet.tflite
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
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 🦊
- 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
