cem

Also, @poor1017 , do you see the same behavior with WSJ0Mix recipe also?

Hi @vishalghor , Thanks for this PR. I can try to review this, this weekend. We can also take the opinions of @popcornell and @mravanelli . But first of all,...

Great, I will take a look today. Thanks @vishalghor !

Hi @vishalghor, thanks a lot for this PR. Code looks great in my opinion. Here are some minor comments before we can merge: - [ ] Currently, we are using...

Hi @vishalghor , I think the most important point is that we should be able to run `train.py` even if there are some missing files. In my case, there were...

Hi @vishalghor , I will give this a review, thank you very much for the updates!

Hi @mravanelli , yes I will take a look into this, hopefully this weekend.

@justinklchan @pplantinga I am not sure what's happening here. I just tried the inference with RE-Sepformer using the latest version of the develop branch of speechbrain, and it is working...

Hi, sorry for my very late reply. Basically if you run the code with `python train.py hparams/sepformer.yaml --data_folder yourpath/wsj0-mix/2speakers --base_folder_dm yourpath/wsj0/si_tr_s --dynamic_mixing True` It will automatically create the resampled (if...

Hi @huangzj421 , sorry for the delay. I will try to do it soon.