Closed CryptoLisboa closed 1 year ago
As of right now, the script that extracts the captions needs to be fixed. This might be because the JSON from which the captions are usually extracted no longer contains the captions.
Same with #37
That's true.
Could https://github.com/openai/whisper be used to extract text from the audio?
In that case, the challenge would be determining which user each sentence belongs to.
As of right now, the script that extracts the captions needs to be fixed. This might be because the JSON from which the captions are usually extracted no longer contains the captions.