WhisperS2T icon indicating copy to clipboard operation
WhisperS2T copied to clipboard

Published 20 hours ago • verified publisher icon shashikg

Language Auto Detection

Open brunjo opened this issue 1 year ago • 4 comments

I noticed that it would fallback to English if no language is specified. Is there a way to automatically predict the language?

brunjo avatar Dec 18 '23 13:12 brunjo

Hi @brunjo, yes, if you don't provide the language, it will fall back to English. I'll try adding that. However, the issue with this approach is that it complicates the batching of segments across multiple files.

shashikg avatar Dec 19 '23 01:12 shashikg

Hi! Any update about this? I think it will need one additional step where we get the language before continuing decoding

AmgadHasan avatar Jul 18 '24 07:07 AmgadHasan

My solution is to refer to how WhisperX does it and then transplant it accordingly.

ustclan avatar Oct 31 '24 08:10 ustclan

@ustclan

can you share your code or the ideas of WhisperX?

twmht avatar Jan 14 '25 12:01 twmht