How to learn the Vits model

#1
by kdrkdrkdr - opened

Blue Archive's data is very small, so how can it be of such high quality?
Is the vits used MB-iSTFT-VITS, not normal?
If not, what model is it?

Yeah, training and usage guide would be much appreciated.

It's just normal VITS but uses a pre-trained model for finetuning. The pre-trained model was trained using a large corpus of Chinese and Japanese. The training guide will be posted soon here.

Can you make Nilou JP ? I don't know how to use this coding thing in android 😅

@sayashi do you have training guide now?

zomehwh changed discussion status to closed

Sign up or log in to comment