grizzled galleon
Model Info :
- Architecture : RVC Arch.
- Vocoder : NSF-HiFiGAN
- Sample Rate : 32kHz
- F0 Extraction : RMVPE
- Embedder : ContentVec
- Pretrain : Legacy Core 1.5
- Precision : FP16 + TF32
Training Details :
- Epochs : 160
- Dataset Length : 5m 10s
- Batch Size : 4
Note :
don’t use this model for high-pitched songs, the dataset only has low notes.
