Metadata-Version: 2.4
Name: cosyvoice2-fr-infer
Version: 0.1.1
Summary: Minimal CosyVoice2 French inference CLI (bundles runtime + Matcha)
Author-email: Tim Luka Horstmann <lukahorstmann@gmx.de>
License: Apache-2.0
Project-URL: homepage, https://huggingface.co/Luka512/CosyVoice2-0.5B-FR
Keywords: tts,text-to-speech,cosyvoice,french,voice-cloning
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Topic :: Multimedia :: Sound/Audio :: Speech
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.9
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: torch>=2.3
Requires-Dist: torchaudio>=2.3
Requires-Dist: onnxruntime>=1.15
Requires-Dist: openai-whisper
Requires-Dist: hyperpyyaml>=1.2
Requires-Dist: huggingface_hub>=0.24
Requires-Dist: WeTextProcessing>=1.0.3
Requires-Dist: inflect>=7.0
Requires-Dist: numpy<1.25,>=1.23
Requires-Dist: tqdm>=4.66
Requires-Dist: soundfile>=0.12
Requires-Dist: transformers==4.40.1
Requires-Dist: diffusers>=0.29
Requires-Dist: modelscope>=1.20
Requires-Dist: omegaconf>=2.3
Requires-Dist: conformer>=0.3.2
Requires-Dist: hydra-core>=1.3
Requires-Dist: einops>=0.6
Requires-Dist: librosa>=0.10
Requires-Dist: scipy>=1.10
Requires-Dist: matplotlib>=3.6
Requires-Dist: rich>=13.0
Requires-Dist: lightning>=2.0
Requires-Dist: rootutils>=1.0
Requires-Dist: gdown>=4.7
Requires-Dist: wget>=3.2
Requires-Dist: phonemizer>=3.0
Requires-Dist: piper-phonemize>=1.0
Requires-Dist: unidecode>=1.3
Requires-Dist: pyyaml>=6.0
Requires-Dist: deepspeed>=0.10
Requires-Dist: pyarrow>=10.0
Requires-Dist: pyworld>=0.3.0
Dynamic: license-file

cosyvoice2-fr-infer
====================

Minimal, plug-and-play CosyVoice2 French inference CLI that downloads the model from Hugging Face and runs cross-lingual cloning. It bundles the required `cosyvoice` runtime and `matcha` module so you don't need the full upstream repo.

## License

This project is licensed under the Apache License 2.0. 

**Note**: This package includes vendored code from:
- [CosyVoice2](https://github.com/FunAudioLLM/CosyVoice2) (Apache 2.0)
- [Matcha-TTS](https://github.com/shivammathur/Matcha-TTS) (Apache 2.0)

All original licenses and attributions are preserved.

Install (editable for local dev)
--------------------------------

```bash
cd standalone_infer
pip install -e .
```

If you are on Linux with GPU, ensure you install torch/torchaudio matching your CUDA and have `onnxruntime-gpu` available. If CPU-only, `onnxruntime` will be sufficient.

Usage
-----

```bash
cosy2-tts \
  --text "Bonjour, je m'appelle ..." \
  --prompt /path/to/prompt.wav \
  --out /tmp/out.wav
```

First run will download the model assets to `~/.cache/cosyvoice2-fr` (configurable via `--model-dir`).

Advanced options: `--setting`, `--llm-run-id`, `--flow-run-id`, `--hifigan-run-id`, `--final`, `--stream`, `--speed`, `--no-text-frontend`, `--repo-id`, `--no-hf`.

Publish to PyPI
---------------

1) Build the wheel and sdist:
```bash
pip install build twine
python3 -m build
```

2) Upload to TestPyPI (recommended first):
```bash
python3 -m twine upload --repository testpypi dist/*
```

3) Upload to PyPI:
```bash
python3 -m twine upload dist/*
```

End-users can then install via:
```bash
pip install cosyvoice2-fr-infer
# GPU users may first install torch/torchaudio from the CUDA index
# pip install torch torchaudio --index-url https://download.pytorch.org/whl/cu121
```


