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
Hi all, I found that when I used the finetuned whisper model (finetuned with whisper-base.en), the transcripts with audio over 30s would repeated one words when I defined only one prompts. I noticed that hallucinations would exist when using whisper large model and VAD could solve it. But these methods did not improve my model. Any suggestions?
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
-
Hi all, I found that when I used the finetuned whisper model (finetuned with whisper-base.en), the transcripts with audio over 30s would repeated one words when I defined only one prompts. I noticed that hallucinations would exist when using whisper large model and VAD could solve it. But these methods did not improve my model. Any suggestions?
Beta Was this translation helpful? Give feedback.
All reactions