You can modify some parameters in the generated config.json and diffusion.yaml diffusion.yaml List of Vocoders 3. Automatically split the dataset into training and validation sets, and generate configuration files. If OnnxHubert/ContentVec as the encoder List of Encoders Optional(Strongly recommend) Optional(Select as Required) NSF-HIFIGAN RMVPE FCPE(Preview version) □ Dataset Preparation □️ Preprocessing 0. If hubertsoft is used as the speech encoder 3. If using contentvec as speech encoder(recommended) 2. □ Model Introduction □ 4.1-Stable Version Update Content □ Questions about compatibility with the 4.0 model □ Shallow diffusion □ Python Version □ Pre-trained Model Files Required 1. The repository and its maintainer, svc develop team, disclaim any association with or liability for the consequences. You bear full responsibility for any problems arising from the usage of non-authorized datasets for training, as well as any resulting consequences. SoftVC VITS Singing Voice Conversion Announcement Disclaimer □ Terms of Use Warning: Please ensure that you address any authorization issues related to the dataset on your own.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |