Clone of the official QuickVC implementation.
Official demo.
main
branch: Refactored & improvedminfast
branch: Fast training with minumum change
Put pretrained model into logs/quickvc
python convert.py
You can change convert.txt to select the target and source
- Hubert-Soft
cd dataset
python encode.py soft dataset/vctk-16k dataset/vctk-16k
- Spectrogram resize data augumentation, please refer to FreeVC.
python train.py
If you want to change the config and model name, change:
parser.add_argument('-c', '--config', type=str, default="./configs/quickvc.json",help='JSON file for configuration')
parser.add_argument('-m', '--model', type=str,default="quickvc",help='Model name')
in utils.py
- Naturalness has Language dependency (c.f. SoftVC) issue#4
- Training time: 1~2week on RTX3090 x1 issue#6
@misc{2302.08296,
Author = {Houjian Guo and Chaoran Liu and Carlos Toshinori Ishi and Hiroshi Ishiguro},
Title = {QuickVC: Any-to-many Voice Conversion Using Inverse Short-time Fourier Transform for Faster Conversion},
Year = {2023},
Eprint = {arXiv:2302.08296},
}
- MS-ISTFT-VITS: Decoder
- Soft-VC: PriorEncoder's Hubert-soft
- FreeVC: data augumentation