How to transcribe multilangual speech properly. Model bevaviour changes unexpectedly. #1588
nathbhargav
started this conversation in
General
Replies: 1 comment 1 reply
-
|
whisper doesnt support multilingual audio |
Beta Was this translation helpful? Give feedback.
1 reply
-
|
Can I make my whisper model to works like that by doing any techniques like fine-tuning and all? |
Beta Was this translation helpful? Give feedback.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
I have a mix speech of "Hindi" and "English". When I passed through the model it should transcribe "Hindi" part in "Hindi" language and "English" part in English language. But whisper model has shown wired behaviour , sometimes model runs all in English, sometimes it translate to english , sometimes it hallucinate in Hindi and sometime it produces Hindi and English both.
How do I control this behaviour??
Beta Was this translation helpful? Give feedback.
All reactions