Whisper Large-v3 Release
![](https://media.kbin.social/media/6d/42/6d425b5b516dd7d24dab67f292ac34d48871911feceae4f5a952a8c532c934e5.png)
![`large-v3` release · openai/whisper · Discussion #1762](https://lemmy.dbzer0.com/pictrs/image/7d5d9ad1-085f-4fa5-a246-a1ec72fa8571.png?format=jpg&thumbnail=256)
github.com
Whisper is a general-purpose speech recognition model. It is trained on a large dataset of diverse audio and is also a multitasking model that can perform multilingual speech recognition, speech translation, and language identification.
The
large-v3
model shows improved performance over a wide variety of languages, and the plot below includes all languages where Whisperlarge-v3
performs lower than 60% error rate on Common Voice 15 and Fleurs, showing 10% to 20% reduction of errors compared to large-v2:
No comments yet. You could be first!