Hi all!
So, somewhat frustratingly, Orca, the screen reader for Linux, doesn't seem to have any kind of OCR support. I deal with virtual machines a lot, and I've had cases where the VM has had no sound. I've been sitting their for ages changing settings in the VM, and even completely destroying and recreating the VM, before I've finally managed to get it to work. With VMWare and NVDA, I can just OCR the screen and find out what the problem is in an instant.
Is there a similar solution for Linux, like an add-on for Orca or something?
FYI, I'm running Fedora 35 with the Mate desktop environment.

There are 2 ways you can do this. You can either take a screenshot of the VM's window with virt-manager's built in functionality, or you can try and get ocrdesktop to work on ubuntu and assign a shortcut via the gnome-control-center to run it. What it'll then do is OCR the current window and give you a caret navigatable text box where you can see what text it picked up.

You mean take a screenshot, email it to myself, download the attachment on my Windows PC and use NVDA to OCR it? Too many houses to go round.

Nope, you can use tesseract and install the english data pack for it via your package manager and use that, or you can use something like ocrdesktop or lios which acts as a frontend.