You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When we use whisperx.load_model(..., device='cuda', device_id=1), we get a quick peek at the wrong (cuda:0) GPU device instead of the desired one. The problem occurs due to different setups for the model and VAD (in the model, we split to device and device_id, but in VAD we use only device name).
When we use
whisperx.load_model(..., device='cuda', device_id=1)
, we get a quick peek at the wrong (cuda:0) GPU device instead of the desired one. The problem occurs due to different setups for the model and VAD (in the model, we split to device and device_id, but in VAD we use only device name).whisperX/whisperx/asr.py
Line 347 in f2da2f8
Bug visually:
Possible solution:
The text was updated successfully, but these errors were encountered: