Failed: Error transcribing audio: Request timed out

Bugs
1

Hey there!

Been getting this error notification recently when I try upload podcast mp3’s. The episodes I want to read/listen to are pretty long; from 1h to 1h30, so I guess that’s why the transcription requests time out.

I want to read AI transcriptions of the episodes while listening, that’s really the big value of Readlang. But I also realised I could use Whisper for the transcriptions, which gives the extra feature of speaker detection. It just helps clear up some of the more conversation’y podcasts like L’Entreprod, which is what I want to listen to. So perhaps Readlang could just facilitate uploads of mp3s and I can paste in my own transcription, thereby circumnavigating the whole transcription thing itself, which is where it’s currently hanging.

Thoughts?

2

Sorry about that.

If you can get hold of, or generate, a transcription in .srt format there is now an alternative method you could try: It’s now possible to upload your own .srt file to sync with an .mp3 file. You should see the option when you start the upload flow whether you start uploading either an .srt or .mp3 file. Please let me know if this works for you or not.

Another thing that might be worth trying is splitting your mp3 into 2 separate smaller files using software like Audacity: https://www.audacityteam.org/, maybe Readlang’s transcription would work better in that case.