nomadkaraoke / python-audio-separator

Easy to use stem (e.g. instrumental/vocals) separation from CLI or as a python package, using a variety of amazing pre-trained models (primarily from UVR)
MIT License
514 stars 86 forks source link

Problem recognizing gpu but only working with cpu #137

Closed kbuwel closed 3 weeks ago

kbuwel commented 1 month ago

conda install pytorch==cuda onnxruntime==cuda audio-separator -c pytorch -c conda-forge 2024-10-17 02:46:58.559 - INFO - separator - ONNX Runtime GPU package installed with version: 1.19.2
10-17 02:46:58.559 - INFO - separator - No hardware acceleration could be configured, running in CPU mode My system is as follows. 5950x cpu, gtx1660 6gb 5600x, rtx3060oc 12gb 2x installed.

beveradb commented 3 weeks ago

Hello! That means you don't have a working CUDA environment.

There's a section in the README about this: https://github.com/nomadkaraoke/python-audio-separator?tab=readme-ov-file#gpu--cuda-specific-installation-steps-with-pip

It's very hardware and system specific so I can't do much to help; though there are plenty of guides online for how to troubleshoot your CUDA environment with PyTorch and ONNXRuntime.

Good luck, I hope you get it working on your machine! 🙏