News for hackers. Stuff that matters.
Re:Wav2vec Overview: Semi and Unsupervised Speech Recognition by jmalicki on Saturday July 03, 2021 @01:10PM (#27723102)
If you look at the architecture diagram for Wav2Vec-U, the "generator" is doing exactly that - generating waveforms from the vectors. All GANs work this way, and is how websites like https://thispersondoesnotexist.com/ work. Of course as the sibling comment notes the results today might not be great for this task, and it is open research, bit it's not as of it just can't be done at all.
Re:Wav2vec Overview: Semi and Unsupervised Speech Recognition by lunixbochs on Saturday July 03, 2021 @01:16PM (#27723142)
My reading of the generator diagram (figure 6) isn't that it is generating waveforms, but that it is generating phoneme probabilities.

You can train a similar system to produce audio on the output of wav2vec, though it probably won't sound similar to the input audio (accent/voice) unless you expose more features of the input than phonemes.