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

How to make the SGMSE in causal way #34

Open
Leaou-Bin opened this issue Sep 21, 2023 · 0 comments
Open

How to make the SGMSE in causal way #34

Leaou-Bin opened this issue Sep 21, 2023 · 0 comments

Comments

@Leaou-Bin
Copy link

Hi, when I reproduce the causality in SGMSE (actual in SGMSE-BBED) as the paper 'SPEECH SIGNAL IMPROVEMENT USING CAUSAL GENERATIVE DIFFUSION MODELS' in ICASSP2023. I make the following changement for score model ncsnpp+ as in paper :

  1. add causal pad for 3*3conv2d.
  2. block the attention module
  3. block the Groupnorm
  4. replace the layerspp.ResnetBlockBigGANpp's downsample_2d, upsample_2d with causal conv2d and TransposedConv2D,block the groupnorm in module
  5. replace the layerspp.Downsample Upsample as above in 4.

But it still not work, could you give me some advice on it, Thanks!

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

1 participant