Hervé BREDIN
Hervé BREDIN
Related: https://pytorch.org/audio/stable/tutorials/audio_resampling_tutorial.html#comparison-against-librosa
Yes. Big RAM, slow I/O as well, here :)
Thinking out loud: would it make sense to make no API distinction between feature extraction and augmentation? Spectrogram extraction could be seen as an augmentation of audio. It would allow...
Related: https://github.com/yluo42/FRA-RIR
[Working on it](https://twitter.com/hbredin/status/1597262288146563072?s=20&t=ckBD2GGTexSYp-p-m9Plxg)
@olvb should be fixed in `develop` branch.
Closing as this will be integrated into [pyannotebook](https://github.com/hbredin/pyannotebook).
Could be related to #1370 if GPU was an Ampere (e.g. A100) device.
[VocalSound](https://github.com/YuanGongND/vocalsound)
[EpicSounds](https://github.com/epic-kitchens/epic-sounds-annotations)