I made a Chinese, Japanese, English model with the VITS-fast-fine-tuning repo, and with that repo, you can specify which language the text is spoken in, Chinese, Japanese, English, or Mix, which mixes all the pronunciations, but it can speak any language. When I use it with Moegoe it automatically uses Mix which is suboptimal. Is there any way to change this?
I made a Chinese, Japanese, English model with the VITS-fast-fine-tuning repo, and with that repo, you can specify which language the text is spoken in, Chinese, Japanese, English, or Mix, which mixes all the pronunciations, but it can speak any language. When I use it with Moegoe it automatically uses Mix which is suboptimal. Is there any way to change this?