A Wen Qingman Qingzhu RVC model

The model's voice is warm and blue. If your voice is strong, you can use the Qing-shu tone, which is decent for singing. Try to lower your voice when using it, and if the effect isn't ideal, you can adjust the pitch to fix it.

It's normal for the displayed reasoning results to differ from those used. Everyone's voice and tone are different, so the results of reasoning will also vary.

The dataset collected 30 minutes of data. To achieve better results, it includes multiple timbres similar to the dataset, such as laughter and screaming, so the tone may be slightly better.

As long as the singing isn't particularly high, it's fine, since this is a warm and youthful model.

Before reasoning

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

After reasoning

Chat effect

Singing performance

parameters

Alchemy log

Note: The dataset comes from the internet, is not copyrighted, and is only for your own research and study. You may delete it within 24 hours after downloading. The fees are limited to the compilation fee, model release fee, sound conversion effect production fee, and video tutorial production fee, and do not cover the model's own copyright fees.