When rendering audio within Voice AI I saw all my 16 AMD Ryzen 9 CPU cores working in a balanced mode so your software has been designed in a muti thteading way.So far this is promising.
Could you please add GPU based rendering into your plugin as this is about factor 20x faster. Use either the CUDA way or by using OpenCL instructions….
4 comments
Hi. Maybe you can tell me why I have never been able to generate through my CPU? Cubase simply crashes. And this is on two computers. Only generation for tokens works. I want to buy a plugin, but if such problems are with local generation, then no
On my Windows 10 Pro system with Cubase 14 Pro I got it to work.
I just opened the plugin did a login within the plugin and registered my perpetual license key and then switched the GUI to “perpetual”.
In Cubase 14 Pro I first wasn't able to get it to work but after reading https://support.sonarworks.com/hc/en-us/community/posts/24795801781266-Bug-in-the-current-Voice-AI-version I was able to capture the original audio and it's fun to use it.
You probably need to create a new posting containing some information about your system and about your DAW and it's version so developers of Sonarworks may see there is an issue that needs to be resolved.
Agree with this feature request. It's 2025 and AI processing on GPU is so well established that I was very surprised to discover that I could not use my GPU and had to rely on my CPU. Come on Sonarworks, there's so many resources available to help run AI models on a GPU. This needs to be a priority.
Asked for GPU support on Windows, still waiting and hoping for it to happen hopefully soon.