Voice changer using high memory and doesnt seem to use my GPU even though its selected

[unkempt hatch]

Any options that I may be missing? Either to reduce memory usage / work harder on the gpu while still producing quality audio
It would work fine being the only thing open (in terms of memory usage of my pc), but I found out that it works horrible or doesnt even work when another application like valorant is open (my memory is 16gb).

Any solutions? Any options I may be missing that could improve it? -Thanks

I'm not sure if I did everything right, as I'm new to this voicechanger

but when adding new voices all I do is upload pth and index files and then tinker with pitch and index

I have no idea what 'export to onnx' is (heard someone mention it)

[crude fern]

Hey there, the realtime voice changer uses quite a lot of your GPU, and as far as I know your current gpu might struggle a lot with running it. And regarding your memory, AI things take up a lot of memory by itself. I would recommend getting more ram and vram to run this.

The best you can do in this situation is to upgrade your ram and possibly video card.

But afaik, yeah you did everything right. Not sure what export to onnx is either but I don't use it.

[near wharf]

Export to onnx is for AMD lol

[near wharf]

just use crepe_tiny and use sup2 only

never use crepe full