Mirco Ravanelli

Yes, it looks like that! On Thu, 29 Aug 2019 at 04:48, Parcollet Titouan wrote: > Hi ! Isn't it simply a problem with the path of the test dataset...

It looks like and error in reading features and labels with kaldi. To debug, you can try to "manually" read the features in this way: 1- select one ark file...

Does /home/spencer/kaldi/egs/timit/s5/data/cmvn_dev.ark exists? Mirco On Thu, 3 Oct 2019 at 09:24, spencerkirn wrote: > Thank you for the quick reply. I apologize if these are basic questions, I > am...

This cmvn file is created by kaldi during the feature extraction phase and it performs mean and variance normalization. You should probably have the cmvn file somewhere else like in...

This is weird, are you sure that you don't have a path problem only? On Fri, 25 Oct 2019 at 08:54, spencerkirn wrote: > In case anyone else has this...

Interesting, we haven't experimented this issue on our side. On Fri, 25 Oct 2019 at 11:07, spencerkirn wrote: > Yes, I checked all the paths in the config file and...

Maybe this file has not been created because there is a problem with test data. Could you better check them? Mirco Virus-free. www.avast.com On Fri, 25 Oct 2019 at 12:19,...

My suggestion is to move to SpeechBrain though: https://speechbrain.github.io/ SpeechBrain is a new project which is much richer, easy-to-use, well-documented than the old pytorch-kaldi one. On Tue, 16 Mar 2021...

Hi Joanna, thank you for your pull request. I think the modification you propose is interesting and can give users more flexibility. This modification requires a change in the config...

Yes, but we are planning to significantly expand this project in the next months with several other functionalities (including some of this pull request). Best, Mirco On Tue, 13 Aug...