@ftshijt, can you answer it?

For training/inference/evaluation, please check https://espnet.github.io/espnet/recipe/svs1.html
For pre-trained models, please check https://huggingface.co/models?other=singing-voice-synthesis&sort=trending&search=espnet

We are also actively working towards an interactive demo (which should be ready in 2 weeks, and will be presented in this year's ACMMM demo, together with https://arxiv.org/abs/2409.07226)

Thank you -_-
I tried to use it but I don't understand the doc🤯 . i have Windows no linux.
There are no pre-trained models: in English, French, Spanish.
I'm waiting for the interactive demo. 😀

Hi!
I'm also interested in this!
Any news on the interactive demo? :)

Hi! I'm also interested in this! Any news on the interactive demo? :)
Hello , here's the demo
https://huggingface.co/spaces/espnet/svs

Many thanks! :)

Thanks for sharing! @funmolde

Yeah, please refer to the model there @migperfer However, the demo was launched in CPU, so it is largely blocked to decoding speed. If you can use GPU for decoding, it would be super fast in general.