dialect-app / dialect

A translation app for GNOME.
GNU General Public License v3.0
596 stars 69 forks source link

Implement OpenAI's Whisper #284

Open m1karii opened 1 year ago

m1karii commented 1 year ago

Whisper is on-premise GPU-based general-purpose multi-task model that can perform multilingual speech recognition as well as speech translation and language identification.

https://github.com/openai/whisper

mufeedali commented 1 year ago

Are you suggesting that we add speech to text?

m1karii commented 1 year ago

no. this AI can do:

i suggest you add the translation part for translating on-premise without the need of internet connection