This version is the RVC version

GPT-SoVITS V2 in: https://apis.klrvc.com/gptsovits/1778.html

This model does not support singing, so it produces a loud electric noise, but it is fine for everyday conversation.

The model is self-trained. I'll make this model available for free download to everyone. As usual, I'll get the download link after commenting

I collected about

20 minutes of dry audio. The sound is very cute, and I hope everyone supports the GPT-SoVITS V2 version. That will be my motivation to keep going.

The effect after reasoning

<!--[if lt IE 9]>document.createElement('audio');<![endif]-->

Screenshot of the training

Download link

Please give my

model a thumbs up. Thank you all. The original model data was collected online and does not have dataset authorization, so it is for study and research purposes only. Any violations are at your own risk.