MaxMax2016
MaxMax2016
Grad-SVC is a diffusion based SVC. So ~ It has no electronic sound, but is sensitive to noise data.
It's not perfect yet. If you are not a developer, you should wait a while until the perfect state is released
so good, 我看了[Multi-GradSpeech](https://arxiv.org/abs/2308.10428)的效果，很好呀，很期望您能开源！
v2 96 with Integrated [Fast Maximum Likelihood Sampling Scheme](https://github.com/huawei-noah/Speech-Backbones/tree/main/DiffVC).I have not test so much. maybe CFM's advantage is speed.
yes, CFM can use less steps.
Different parameters have different effects, such as big model gets better result. Grad-SVC & so-vits-svc 5.0 are all just demo with small model for SVC, Their true abilities have not...
https://www.zhangxueyao.com/data/MultipleContentsSVC/index.html some one else may will do.
train by the open source data：https://github.com/Multi-Singer/Multi-Singer.github.io
The versions used during the development don't mean that only these versions can be used. The new versions haven't been verified yet.
pip install WeTextProcessing