|
|
||
|---|---|---|
| realtime_websocket | ||
| scripts | ||
| tests | ||
| .env.example | ||
| .env.funasr.example | ||
| .env.qwen_legacy.example | ||
| .gitignore | ||
| FUNASR_README.md | ||
| README.md | ||
| README_QWEN_LEGACY.md | ||
| model_manifest.json | ||
| pyproject.toml | ||
| pyproject_qwen_legacy.toml | ||
| requirements-auxiliary.txt | ||
| requirements-funasr.txt | ||
| requirements-qwen-legacy.txt | ||
README.md
FunASR realtime ASR demo
This branch runs streaming ASR, VAD, and CAM++ from local model assets. The browser UI starts as a separate service.
Start
Install a torch/torchaudio build for the host, then install requirements-funasr.txt and requirements-auxiliary.txt. Copy .env.funasr.example to .env if no .env exists, then download the three required FunASR models with the shared project manifest:
python scripts/download_models.py --funasr-runtime
Start the backend (CAM++ model service plus WebSocket) in one terminal:
python scripts\run_backend.py
Start the frontend in another terminal:
python scripts\run_frontend.py
Open http://127.0.0.1:8080/. Model directory layout and address settings are described in FUNASR_README.md.