Map screen 1 to audio 2 for interpreting

I would like my audience to see a presenter’s screen while listening to the audio stream of one or more interpreters. The interpreter can do a screenshare of the presentation but this lowers the video quality as it’s a 2nd-generation stream. However, ideally, it would be the interpreter controlling this so they could switch to visuals/screens in their specific language and their audience would be tuned in only to the interpreter’s broadcast
In other words, interpreter broadcast of presenter’s screen would not be a download>re-upload ( multiplied by x interpreters) but a direct mapping across to the presenter’s visual in substitute of their own.
If this is possible it would open a huge scope of possibilities. Is it?

Here’s a different approach that I think gives you the same results: Adjust volume levels via javascript - #15 by izakgl (but it still requires a pull request to be merged).