You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Would love to add support for this feature request - it's incredibly helpful for the transcription of long audio files, such as podcasts, in order to help whisper get the spelling of names (people, places, companies, etc.) right
Whisper recently added an option to carry the initial prompt with the sliding context window: openai/whisper@5979f03
As far as I can tell, there's no way to set this option to True in whisper.cpp?
The text was updated successfully, but these errors were encountered: