News for hackers. Stuff that matters.
Re:StyleTTS2 – open-source Eleven Labs quality Text To Speech by modeless on Sunday November 19, 2023 @08:27PM (#38340938)
Great question! Whisper processes audio in 30 second chunks. But on a fast GPU it can finish in only 100 milliseconds or so. So you can run it 10+ times per second and get around 100ms latency. Even better actually because Whisper will predict past the end of the audio sometimes.

There are no comments on this comment yet.