how to replace whisper model to increase precision?

alphacep / vosk-api

Offline speech recognition API for Android, iOS, Raspberry Pi and servers with Python, Java, C# and Node

Apache License 2.0

7.63k stars 1.07k forks source link

Open laosuan opened 1 year ago

laosuan commented 1 year ago

Recently, OpenAI released https://github.com/openai/whisper, which had good performance. how to replace whisper model to increase precision?

nshmyrev commented 1 year ago

Hi. We are working on that feature, things in progress. Stay tuned.

Btw, for many applications (callcenter, bad audio quality) the difference between Whisper and Vosk models is minimal. You can check in this page: