News for hackers. Stuff that matters.

A comment on Nvidia Accelerates Real Time Speech to Text Transcription 3500x with Kaldi.

Re:Nvidia Accelerates Real Time Speech to Text Transcription... by melling on Monday March 18, 2019 @03:59PM (#19424605)
“This means 24 hours worth of human speech can be transcribed in 25 seconds.”

With this kind of performance, can’t a large body of video or audio (e.g podcasts) be transcribed then manually corrected to improve the model?

[ Parent | on HN ]
Re:Nvidia Accelerates Real Time Speech to Text Transcription... by intopieces on Tuesday March 19, 2019 @12:01AM (#19427986)
A little bit, but podcasts and other instances of performed speech represent a narrow sample of what human speech sounds like in the real world. Eventually, there is not much to gain from it.
[ Parent | on HN ]