Nabarun Goswami

Is there a concrete example of using tensorcom and webdataset?

1

I am trying to create a distributed preprocessing pipeline using tensorcom. I have successfully created it and I can access the data from a different machine. However, I am trying...

Hi, May I know what was the training hardware details for the experiments in the paper. Specifically, - How many GPUs were used? - What was the batch size per...

Hi, I am wondering what are your parameters for extracting the mel spectrograms. And also, if you use the log-mel spectrogram as target for training the model?

Added _read_mp4box_stem_titles function

3

Basically, for stems which have been created by the NI creator have the titles of the streams in the `udta` metadata. I added the functionality to try to read the...

Linux version of MEX files

1

Hi, could you provide linux builds of the pesq mex functions? Regards Nabarun

Could you add a license?

1

Hi, could you add a license or terms of use for the repository and pre-trained models?

`WaveformToFbankConverter` does not work as expected for `channels > 1`

2

**Describe the bug:** While the `AudioDecoder` can read a stereo wav file, the output of `WaveformToFbankConverter` is always `num_frames x num_mel_bins`. **Describe how to reproduce:** ```python audio_wav_path, device, dtype =...