Zero-pad the audio, not the spectrogram

[ggerganov]

Makes sense - hopefully will reduce hallucinations

https://github.com/openai/whisper/discussions/838#discussioncomment-5222689

[lunixbochs]

(Removed previous incorrect assumption about the featurization).

As long as you don't change the featurization, it looks like you can just switch your padding to -1.5 if it's more convenient to keep padding at the spectrogram level.

It seems like your speed_up could interfere with the padding otherwise

[meakbiyik]

Oh this might be a game-changer, I wasn't able to find a way to reduce them so far.

[albino1]

This was merged into Whisper as of 20230307, is there a chance we'll see it in whisper.cpp soon?