Hacker News Reader

I gave it a spin a little bit ago. Per usual, install docs didn't quite work OOTB, here's how I got it working: https://llm-tracker.info/books/howto-guides/page/speech-to-t...

One limitation that seems undocumented, the current code only supports relatively short clips so isn't suitable for long transcriptions:

> ValueError: The input sequence length must be less than or equal to the maximum sequence length (4096), but is 99945 instead.

Seems like you could easily do a little bash/python script to chop up the recording and batch process each, then stitch the results together?

Probably, although you could more easily use WhisperX and get the same results twice as fast and without any additional scripting.

Does WhisperX do translation? The repo suggests it's only for TTS, timestamps and diarization.

Yes, just use the "translate" instead of the "transcribe" option.