https://github.com/kahsolt/soft-vc-realtime
A wrapper for Soft-VC to make a real-time voice converter.
https://github.com/kahsolt/soft-vc-realtime
Last synced: 11 months ago
JSON representation
A wrapper for Soft-VC to make a real-time voice converter.
- Host: GitHub
- URL: https://github.com/kahsolt/soft-vc-realtime
- Owner: Kahsolt
- License: mit
- Created: 2022-11-02T13:02:00.000Z (about 3 years ago)
- Default Branch: main
- Last Pushed: 2022-11-02T13:03:19.000Z (about 3 years ago)
- Last Synced: 2025-01-03T16:28:47.246Z (about 1 year ago)
- Language: Python
- Size: 3.91 KB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# soft-vc-realtime
A wrapper for Soft-VC to make a real-time voice converter.
----
### Quick Start
- install PyTorch following the official guide: [https://pytorch.org/get-started/locally/](https://pytorch.org/get-started/locally/), only need **torch** and **torchaudio** parts
- install dependecies `pip install -r requirements.txt`
- run the wiring test `python app.py --mode echo` to check your audio device settings ok
- now, run in default vc mode `python app.py`
#### trouble shooting
Q: When running in vc mode, there's no GUI windows and it directly quits after model loading
A: Open and edit file `%USERPROFILE%\.cache\torch\hub\bshall_hifigan_main\hifigan\utils.py`, comment the 4-th line `matplotlib.use("Agg")`
----
by Armit
2022/11/02