The super cute Ani loli voice GPT-SoVITS V2

This is my first time posting my model on Miaoyin. Thanks to Carrier for their guidance

Ani's version is divided into RVC and GPT-SoVITS. The current model you see is generated by GPT-SoVITS V2. A reminder to everyone who purchased it: the model cannot be used commercially.

The model used about 5 minutes of dataset, and the dataset effects were edited frame by frame by myself, so it was a bit tough, but the results were excellent.

I'm pretty lazy, so I didn't create reference audio, but all the reference audio was manually split by me and is usable, essentially splitting the entire dataset.

Training sessions were 10 and 15 rounds respectively.

After reasoning

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

After some reasoning

,

Screenshot

It should be noted that GPT-SoVITS has higher timbre accuracy than RVC. Purchasing does not mean you own the timbre; you are only paying for the model's service fee. The downloaded model is only for your research and should not be used for commercial purposes, or you will bear the consequences.