Why is there a need for Torch implementation of STFT and MelSpectrogram?

[brihijoshi]

Hi Jong,
Thank you for such a useful implementation! Sorry for a silly doubt, but I am a beginner in MIR and working with Onsets and Frames for a project of mine.

Looking at the STFT and Mel Spectrogram classes, it doesn't seem (and there possibly isn't a need) that they are 'learnable'. I am a little confused as to why is there a need for a Torch implementation for STFT and Mel Spectrogram? Is it not possible to use librosa implementations for the same?

Thanks!

[jongwook]

Hi, those were in order to perform Mel spectrogram calculation in GPU, rather than in CPU (librosa), hoping it to be faster. Ase you said, the filters in those modules are not learnable because we used register_buffer instead of nn.Parameter. Those were adapted from NVIDIA's Tacotron2 code, and similar modules exist in torchaudio.