2 ms·

Can anyone confirm/deny my theory that AI audio generation has been lagging behind progress in image generation because it’s way easier to get a billion labeled

• madelyn 4y ago
  Sound is a lot higher fidelity, it's harder to make the information available to a computer without serious downsampling or simplification. Consider sounds over 12khz. On a spectrogram during a chorus or drop that area is lit up, with so many things changing from millisecond to millisecond. A lot of AI samples really struggle at high frequencies, or even forgo them entirely. Midi based approaches have been really great though, and an approach like in the OP is fascinating (and impressive).
• antognini 4y ago
  As someone who works in the audio AI space I think an underappreciated reason for audio lagging behind is that it's a lot harder to put impressive audio clips in an academic paper than impressive images.