Max4Live Devices from AI Prompts

Misc
Blush Audio's icon

Blush Audio

Hello,

I run a module at a university that uses Max4Live to implement key aspects of Digital Signal Processing.

However, with it being a visual programming language it's not very accessible should we ever have visually impaired students on the module. So with that in mind, I was wondering if it's possible to create M4L devices with AI prompts.

I've had success in Gemini spitting out some JSON, but I can't get it to work either by popping it in an .amxd (trying to preserve the file header), or by pasting in as text.

Has anyone ever tried this? I'm using Gemini Pro btw.

Thanks!

Source Audio's icon

Source Audio

Same as with mxf files, amxd header is not only describing device kind

but also has checksum that gets calculated from various factors.

That is why pasting json text into amxd can't work.

but if that json files are ok as max patches,

you could paste the code into empty max editor patch in Live and

then save it as amxd.