https://github.com/seanghay/uvr
Ultimate Vocal Remover CLI
https://github.com/seanghay/uvr
uvr vocalremover
Last synced: 2 months ago
JSON representation
Ultimate Vocal Remover CLI
- Host: GitHub
- URL: https://github.com/seanghay/uvr
- Owner: seanghay
- Created: 2023-06-13T12:25:32.000Z (about 2 years ago)
- Default Branch: main
- Last Pushed: 2025-02-05T03:22:37.000Z (4 months ago)
- Last Synced: 2025-03-29T08:03:10.055Z (3 months ago)
- Topics: uvr, vocalremover
- Language: Python
- Homepage:
- Size: 4.45 MB
- Stars: 137
- Watchers: 1
- Forks: 32
- Open Issues: 4
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
📦 Check out my new library called [Vocal](https://github.com/seanghay/vocal) for **Vocal/Speech Separation** built with simplicity in mind.
---
# Ultimate Vocal Remover CLI
[[Colab]](https://colab.research.google.com/drive/1VDncdndceKanFrs2LU-LM4Odv8tnPkzD?usp=sharing)
⚠️ Before running this project, make sure you have installed `torch`, `torchaudio`. Please check out the PyTorch documentation.
⚠️ Also make sure you have `libsndfile` and `ffmpeg` installed.
⚠️ This project currently works on CUDA.
## Install dependencies
```shell
pip install -r requirements.txt
```
## Download Model Weights
```shell
./download.sh
```
## Separation
```shell
python separate.py
```
[Modify these lines to fit your needs](https://github.com/seanghay/uvr/blob/fa19a9821d42586883202623936a0c8b895ae047/separate.py#L101-L108)
```python
if __name__ == '__main__':
device = 'cuda'
is_half=True
model_path='uvr5_weights/2_HP-UVR.pth'
pre_fun = _audio_pre_(model_path=model_path,device=device,is_half=True)
audio_path = 'audio.aac'
save_path = 'opt'
pre_fun._path_audio_(audio_path , save_path,save_path)
```