FFmpeg could not read file and torchaudio could not open seperate.

[constantinethegr8]

❓ Questions

I use windows and have installed ffmpeg at C:\ffmpeg\bin. I even have it in environment variables. I have never had problems like this before and most seperations would happen smoothly. I tried updating with python.exe -m pip install -U demucs but I'm apparently up to date. I have never needed to type python -m when using demucs. Even when I did, the seperation would not work. I think ffmpeg could be the problem. I highly doubt pytorch had much to do with it. Here's my log:

C:\Users\username\Desktop\youtube-dl>python -m demucs separate "That's What I Like-JucvYrdSIcM.wav" -d cuda -o separate --v
erbose
Important: the default model was recently changed to `htdemucs` the latest Hybrid Transformer Demucs model. In some cases, this model can actually perform worse than previous models. To get back the old default model use `-n mdx_extra_q`.
Selected model is a bag of 1 models. You will see that many progress bars per track.
Separated tracks will be stored in C:\Users\username\Desktop\youtube-dl\separate\htdemucs
Separating track separate
Could not load file separate. Maybe it is not a supported file format?
When trying to load using ffmpeg, got the following error: FFmpeg could not read the file.
When trying to load using torchaudio, got the following error: Error opening 'separate': System error.

[CarlGao4]

You should run python -m demucs.separate "That's What I Like-JucvYrdSIcM.wav" -d cuda -o separate --v instead.