Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

训练没效果 #93

Open
endlllll opened this issue Oct 9, 2023 · 4 comments
Open

训练没效果 #93

endlllll opened this issue Oct 9, 2023 · 4 comments

Comments

@endlllll
Copy link

endlllll commented Oct 9, 2023

我使用 4.0 训练了 10k 和 100k 后进行了对比，转换出来的音频没有任何差异，与目标音色差距也非常大。

训练过程均使用的默认配置与默认的预训练模型，没有做任何改动。

image
@yxlllc
Copy link
Owner

yxlllc commented Oct 9, 2023

这过拟合了吧，数据太少或者检查一下数据集是否都是相同音色的

@endlllll
Copy link
Author

endlllll commented Oct 9, 2023

我用了 2 个小时时长的说话干声训练的，理论上数据量足够了，数据集也都是相同音色的
同样的数据集使用 RVC 训练过一次，效果还 OK

@yxlllc
Copy link
Owner

yxlllc commented Oct 9, 2023

音频切片时间小于2秒的会被直接忽略。反正你这个验证集loss肯定是有问题的

@y010204025
Copy link

自己推理点歌曲试试吧，以听感为主

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@endlllll @y010204025 @yxlllc