m-bain / whisperX

WhisperX: Automatic Speech Recognition with Word-level Timestamps (& Diarization)
BSD 2-Clause "Simplified" License
11.39k stars 1.2k forks source link

Load Model To CPU and Then GPU #809

Open twicer-is-coder opened 3 months ago

twicer-is-coder commented 3 months ago

Anyway to load the model in CPU first and then move to GPU?

tophee commented 3 months ago

To use CPU, set device = "cpu". Once you want to switch to GPU, set device = "cuda"