GitHub - openai/whisper: Robust Speech Recognition

Elm Radio is using this to automatically generate transcripts of their podcast.

Here is what Dillon has to say about it:


[00:29:18]
Another thing that I shipped this year that I’m really excited about was OpenAI Whisper

[00:29:24]
is this voice transcription service that works very well for Elm Radio.

[00:29:29]
I tried several paid and free transcription services and they were pretty bad.

[00:29:34]
Half a year ago, a year ago, even the paid ones, there were some things I’m like, I definitely

[00:29:38]
don’t want to put that up in a public place because that’s a very embarrassing typo in

[00:29:43]
the transcript.

[00:29:45]
And then I tried OpenAI Whisper, this recently released like MIT licensed transcription project.

[00:29:52]
And it’s amazing.