News for hackers. Stuff that matters.
Re:A speech-to-text practitioner’s criticisms of industry an... by est31 on Monday April 06, 2020 @08:59AM (#22793294)
> Audiobooks could be another resource

Yes, that's in fact the source of the commonly cited LibriSpeech dataset. It used publicly available recordings from the LibriVox project and has done some cleaning steps as well as alignment of the transcript with the audio to arrive at 1000 hours of cleaned up audio.

The M-AILABS dataset uses LibriVox as well, among other sources, to arrive at 1000 hours of cleaned up data in various european languages.

Overall there's still a large untapped potential in LibriVox data: https://gist.github.com/est31/1e195c55fab8f95a72393db1519da1...

But you'll have to give up properties like gender balancing in your dataset that then has to be counteracted during the learning process, e.g. by having gender-custom learning rates.

Re:A speech-to-text practitioner’s criticisms of industry an... by lunixbochs on Monday April 06, 2020 @11:46AM (#22794716)
Facebook’s librilight taps this potential. They have a script for easily fetching an entire language from librivox, and I have a pipeline on top of that for book fetching and transcript alignment that I’ve successfully trained from.

I think the next step here needs to be auto-finding librivox source books that aren’t in e.g. Gutenberg. I only auto-found books for 75% of English. Only found books for 20% of German. I don’t actually know where to look beyond Gutenberg for books that don’t have easily linked text. Common crawl? Librivox forums?

(Or you can take Facebook’s semi-supervised approach and generate a transcript from thin air instead, but imo that will work better for English than languages that don’t have good baseline models yet)