Hacker News Reader

WER breakdown by languages of Fleurs dataset, using the large model: https://github.com/openai/whisper/blob/main/language-breakdo...

from https://github.com/openai/whisper#available-models-and-langu...

I've got people to test and usually with the 'medium' and 'large' models it works really efficiently. Honestly, I just use the large model for everything, because might as well have the best quality if you're going to do the effort.

I've been using it on lots of Russian stuff and it's great, even translates for you.

I’ve tried it on French and results were pretty good

Spanish is great.