We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
请问在代码里哪里体现参考NaturalSpeech的损失函数了?
The text was updated successfully, but these errors were encountered:
https://github.com/PlayVoice/vits_chinese/blob/master/models.py#L549~L550 https://github.com/PlayVoice/vits_chinese/blob/master/train.py#L264
Sorry, something went wrong.
好的,谢谢
@MaxMax2016 再问下,加和不加这个损失,做过对比吗?效果差异明显吗?
在标贝数据集上,发音错误少些;VITS的发音错误一直被诟病。
No branches or pull requests
请问在代码里哪里体现参考NaturalSpeech的损失函数了?
The text was updated successfully, but these errors were encountered: