Problems with Kaldi MFCCs

Hi,
thank you very much for this very useful project.

I started doing some speech recognition experiments with the MFCC features implemented in torchaudio. In particular, I tried the librosa ones implemented in torchaudio/transorms.py and the kaldi-ones implemented in torchaudio/compliance/kaldi.py.

• The librosa features are computed very efficiently and I can achieve results similar to that of the original kaldi features when changing some hyperparameters (i.e, n_mfcc=13, hop_length=160,n_mels=23,f_min=80,f_max=7900).

• When switching to the kaldi implemented features, however, my neural network doesn't even converge. I suspect there a bug somewhere. I tried to compare the original kaldi mfccs with the ones implemented in torchaudio and they look very different (dithering only cannot explain such a big difference):

mfcc_original
array([35.84189 , 39.748493, 35.40782 , 33.237488, 34.53969 , 35.40782 ,
       34.973755, 35.40782 , 35.40782 , 35.84189 ], dtype=float32)

mfcc_torch
tensor([29.3794, 29.1657, 28.7020, 27.4892, 29.1944, 27.8915, 29.3321, 28.8958, 28.4197,
29.0967])

The other issue is that the current version doesn't support cuda and it can only process up to two-channels at a time. Also, the current version is significantly slower than the librosa implementation (there could be a bottleneck somewhere).

Any idea?
Hope my feedback would be helpful

Thank you

Mirco

Start by reproducing the reported MFCC comparison using torchaudio/transorms.py and torchaudio/compliance/kaldi.py, then compare the results with the original Kaldi output. Investigate the reported CUDA, channel-count, and performance limitations as separate concerns. Done means identifying the source of the discrepancy and defining verified behavior for the affected cases.