> A possible option may be to run the screen reader on the remote desktop itself if that's possible. For generic remote desktop access, that's what we have to

> A possible option may be to run the screen reader on the remote desktop itself if that's possible. For generic remote desktop access, that's what we have to do. But piping accessibility information (in its generic form, not pre-rendered as speech) to the client side in that case is much harder than it would be for this Chromium-based remote browser. (I know this from relevant experience during my time on the Windows accessibility team at Microsoft.)