PyPI page
Home page
Author:
Warith HARCHAOUI
License:
BSD-3-Clause
Summary:
Vocal Helper — async producer/consumer pipeline turning a live PCM stream into diarized, transcribed utterances and (optionally) a rolling LLM summary. Stages: Voice Activity Detection → online speaker diarization (Hungarian-style cosine clustering on backend embeddings) → STT (pywhispercpp turbo) → optional analyst (a local model served via Ollama or vLLM, resolved by best-engine-ai-helper from a committed brief).
Latest version:
2.0.5
Required dependencies:
audio-helper
|
best-engine-ai-helper
|
numpy
|
os-helper
|
pywhispercpp
|
scipy
|
silero-vad
Optional dependencies:
best-engine-ai-helper
|
capture-helper
|
click
|
fastapi
|
fastapi-mcp
|
httpx
|
nemo-toolkit
|
ollama
|
podcast-helper
|
pyannote-audio
|
pytest
|
python-multipart
|
sherpa-onnx
|
speechbrain
|
torch
|
torchaudio
|
uvicorn
|
vocal-helper
Downloads last day:
6
Downloads last week:
74
Downloads last month:
190
7-day smoothing