Quick context:
I’m running Whisper and Pyannote Audio diarization in a google cloud VM instance group. Whisper is able to detect my GPU just fine, but Pyannote cannot. torch.cuda.is_available() returns true.
Below describes my setup, which should make the issue reproducible. I have tried to use the development branch, torch 2.0.0, upgraded cuda, etc. Note: Pyannote worked for me in the past, not sure what happened.
Running the following txt file:
torch==1.11.0
torchaudio==0.11.0
torchvision==0.12.0
pyannote.audio
Docker file:
FROM nvidia/cuda:11.2.1-runtime-ubuntu20.04
RUN apt-get update && apt-get install -y --no-install-recommends
python3.9
python3-pip
&&
apt-get clean &&
rm -rf /var/lib/apt/lists/*
Followed these directions for starting the VM with GPU: