Aaron Boxer 18b897277f Add real-time speaker diarization support ...

- New whisper_live/diarization.py: SpeakerDiarizer with online clustering
- Uses pyannote.audio speaker embeddings (optional dependency)
- Cosine similarity threshold for speaker matching (default 0.55)
- Running average embedding update for speaker stability
- Configurable max_speakers limit (default 10)
- Client options: enable_diarization, max_speakers
- Segments include 'speaker' field when diarization is active
- Graceful fallback: logs warning if pyannote not installed
- Added 12 unit tests (mock-based, no GPU required)

2026-05-13 10:50:59 -04:00

..

backend

Add real-time speaker diarization support

2026-05-13 10:50:59 -04:00

transcriber

update to support faster whisper 1.2.0

2025-10-07 14:20:20 +00:00

__init__.py

Expose __version__ in package root and update dependencies in setup.py

2026-03-17 13:34:34 +05:30

__version__.py

Bump version v0.8.0

2026-03-17 14:48:07 +05:30

batch_inference.py

Fix NoneType crash in _process_single when VAD filters all audio

2026-03-11 16:48:23 +00:00

client.py

Add real-time speaker diarization support

2026-05-13 10:50:59 -04:00

diarization.py

Add real-time speaker diarization support

2026-05-13 10:50:59 -04:00

metrics.py

feat: add Prometheus metrics instrumentation

2026-05-13 10:45:40 -04:00

server.py

Add real-time speaker diarization support

2026-05-13 10:50:59 -04:00

utils.py

Replace os.system() in clear_screen() with ANSI escape codes

2026-04-17 09:31:00 -04:00

vad.py

Upgrade silero vad v5.0

2025-01-13 11:37:58 +00:00