Open MituButChi opened 2 months ago
In your case I would:
pip install
& conda install
.
In your case I would:
Setup the environment using python virtual environments to avoid conflicts by mixing
pip install
&conda install
.
- Then call pip cache purge.
- Call whisperX using "small" model instead of e.g. "large-v2" in case it meets the expectations.
Hi. Thank you for your response.
I thought the whole thing took ~10Gb, turned out it took ~25Gb :( Thanks to you I got back ~5Gb. I really appreciate your help.
Hello everyone. This is not an issue but I don’t know where else to post so I hope it’s okay. And I pretty much know nothing about Python or coding etc. so please bear with me :(
I use Windows 10, and here’s the steps that I did:
conda create --name whisperx python=3.10
conda activate whisperx
conda install pytorch==2.0.0 torchaudio==2.0.0 pytorch-cuda=11.8 -c pytorch -c nvidia
pip install git+https://github.com/m-bain/whisperx.git
I will use WhisperX with Subtitle Edit, not with command lines, just in case it helps.
The whole thing takes up more than 10Gb so I do hope I can get back some, but in case it’s not possible I’ll accept my fate.
Thank you so much 👏🏻