SYSTRAN / faster-whisper

Faster Whisper transcription with CTranslate2

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

to_cpu isn't switching

etdapt opened this issue · comments

to_cpu = self.model.device == "cuda" and len(self.model.device_index) > 1

I don't have CUDA, and when this line was encountered, to_cpu stays False

to_cpu=False will have no effect on CPU usage. If you want to run on GPU, you will have to additionally install cuBLAS and cuDNN for CUDA.