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

杂音很重 #55

Open
cooldoomsday opened this issue Apr 26, 2023 · 10 comments
Open

杂音很重 #55

cooldoomsday opened this issue Apr 26, 2023 · 10 comments

Comments

@cooldoomsday
Copy link

按照步骤训练出来的生成杂音很重，会是什么原因
@MaxMax2016
Copy link
Collaborator

训的48K么？现在48K唱低音是有问题，不知道您遇到的是不是这样的问题？

@cooldoomsday
Copy link
Author

16k的按照中文的文档来操作的，整个生成的大部分都是杂音

@cooldoomsday
Copy link
Author

svc_out(1).wav.zip
这是生成出来的

@MaxMax2016
Copy link
Collaborator

16K的模型要用16K的代码，https://github.com/PlayVoice/lora-svc-16k 这里有备份，您看看是不是16K的代码呢？

@MaxMax2016
Copy link
Collaborator

预训练模型是可以直接推理的，可以用预训练模型试试，需要设置发音人config/singers/****.npy

@cooldoomsday
Copy link
Author

预训练模型可以的，那是不是我训练步骤有问题

@MaxMax2016
Copy link
Collaborator

是不是内容编码和提取的pitch不匹配呢？svc_out_pitch.wav正常不？

@cooldoomsday
Copy link
Author

svc_out_pitch.wav正常的

@MaxMax2016
Copy link
Collaborator

试试用tensorboard --logdir logs/ 看看训练过程中生成的音频正不正常

@cooldoomsday
Copy link
Author

不正常，奇怪了，都听不太见人声，电音加杂音

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
2 participants
@MaxMax2016 @cooldoomsday