We just open sourced an incredible speech recognition model. It's generally better than me, and works incredibly robustly across background noise, accents and language mixtures, which is really exciting! Let's use it everywhere!
Conversation
Replying to
It works in any language which is incredible. It can also transcribe any language to English!
2
1
38
Replying to
Right in time! Just Yesterday, I was comparing speech-to-text solutions. Thanks, guys! :)
1
Replying to
Great work! I look forward to trying these out. How easy would it be to use this near-realtime? I assume you'd need some tricks with a sliding 30s window and would probably need to stick with the smaller models?
6
Replying to
Great! I would like to fine tune with spanish language.
2
Replying to
Looks very impressive! Working with 680,000 hours of data sounds like just a little bit of pain. 😅
3
Replying to
hi Boris, I'm wondering what the current accuracies are for various accents, and how those have been changing over time. Is there an easy pointer to this info, and to how Whisper does on this? (have quickly skimmed paper but not dug into it, is it in there?)
2