Chenda Li

Results 10 comments of Chenda Li

> Hi everyone, thanks for this great resources. I wanted to know where could I find the different recipes for speech enhancement (for single source). I can't find them on...

> Should I deduce from this quote that no pretrained models are available for speech enhancement at the moment ? (I'm looking for a SE denoising model :) ) At...

Some parameters like the `weight` $\gamma$ in `BatchNorm`, the `weight` $\alpha$ in `PReLU`, and some custom parameters defined with `torch.nn.Parameter()` may make `p.dim() == 1` be `True`. One possible solution...

BTW, this function is called by amlost all the tasks in ESPnet2, I'm not sure if these updating keeps the previously trained model reproducible.

> > BTW, this function is called by amlost all the tasks in ESPnet2, I'm not sure if these updating keeps the previously trained model reproducible. > > It should...

The PR is merged, so I close this issue.

I have the same question after reviewing these lines of code. @NguyenHai7120 Have you figured it out?

> I understand that you tried to separate this PR and the recipe PR #5810 If you fix the CI error, please let me know. > > @LiChenda, can you...

Thank you @wendlerc for pointing out the issue. I have opened a [PR](https://github.com/espnet/notebook/pull/46) to fix it.