Directory › Audio & voice › OmniVoice
> GEM AUDIO & VOICE · LOCAL AI
OmniVoice
by k2-fsa
Clone a voice from a few seconds of audio, in 600+ languages: the local alternative to ElevenLabs.
View on GitHub →What it does
- Training-free voice cloning, or design a voice by its attributes (age, pitch, accent)
- 600+ languages including French; non-verbal sounds like [laughter] and pronunciation fixes
- Python, command line, or an online demo on Hugging Face
! The code is Apache-2.0, but the model weights' licence is not stated on its Hugging Face card: check it before any commercial use. Only clone your own voice or that of a consenting person.
Install it
Requirements: Python with PyTorch (the right index for your graphics card).
PyTorch for NVIDIA (CUDA 12.8)
pip install torch==2.8.0+cu128 torchaudio==2.8.0+cu128 --extra-index-url https://download.pytorch.org/whl/cu128Install OmniVoice
pip install omnivoice
The prompt for your agent
Paste it into your coding agent: it checks the requirements and installs it for you.
Claude CodeCodexCursorGemini CLIopencode
Prompt
Install OmniVoice (https://github.com/k2-fsa/OmniVoice) on my machine.
1. Check the requirements: Python with PyTorch (the right index for your graphics card). If anything is missing, tell me before installing anything.
2. Follow the official method:
- PyTorch for NVIDIA (CUDA 12.8): pip install torch==2.8.0+cu128 torchaudio==2.8.0+cu128 --extra-index-url https://download.pytorch.org/whl/cu128
- Install OmniVoice: pip install omnivoice
3. Check that it works, then explain how to launch it and where to start.
Always review what your agent is about to run, especially install commands.