Running into this error when trying to generate the dataset
ycemsubakan opened this issue · 3 comments
ycemsubakan commented
JorisCos commented
Thanks for reporting this issue.
I think you tried to run augment_train_noise.py
multiple times.
The original sounds in wham noise tr have 2 channels the augmented one have only one.
If you have 60 000 files in wham noise tr you should skip the augmentation part.
If you don't I suggest that you start from a fresh wham noise tr folder where there are no augmented files and run the script again .
JorisCos commented
Any news @ycemsubakan ?
ycemsubakan commented
Sorry, I have decided not to use the dataset for the time being. I'll let you know if I run into issues again. But it sounded like this was an error on my side. We can close the issue. Thanks!