linto-ai / whisper-timestamped

Multilingual Automatic Speech Recognition with word-level timestamps and confidence
GNU Affero General Public License v3.0
2.01k stars 156 forks source link

I want to get high performance - azure #90

Closed rairavi closed 1 year ago

rairavi commented 1 year ago

suggest me on which server kind we need to deploy this model on azure stack to get high performance, for large-v2 model and long videos?

Jeronymous commented 1 year ago

Please have a look at the discussions of whisper's repo for that https://github.com/openai/whisper/discussions (and open a discussion if you don't find your answer).