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.

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

top10 acc for ar? #8

Open
yuzuda283 opened this issue Sep 20, 2024 · 1 comment
Open

top10 acc for ar? #8

yuzuda283 opened this issue Sep 20, 2024 · 1 comment

Comments

@yuzuda283
Copy link

Thanks for opensource the excellent work!
I was trying to train an AR(GPT) model on dac and xcodec with libritts, in my exp, the top10 acc for the first token of dac and xcodec is very close, around 50%~60%, is that reasonable?
Intuitively,the acc of xcodec should be higher.

@zhenye234
Copy link
Owner

Thank you for your interest in our work and for conducting experiments with it!

Based on your results, the accuracy rates for the first token of dac and xcodec being close to 50% to 60% seem reasonable. Because we integrate both acoustic and semantic information to enhance model performance. From previous experiences, during the autoregressive (AR) stage of the training model Valle, we observed top-10 accuracy typically ranging from 60% to 70%.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants