Maybe it’s already possible to do, but if not here is the idea :
Allow any user to record separately all other users to a separate audio files
or (much better but much more complex):
Provide a way to have separate audio streams available somehow from the server, and find a way to have access to those as a virtual sound card, one per channel. This would allow to use jitsy as a realtime remote audio studio solution. There probably IS a market for that 
(especially in those covid times). The existing solutions looks crappy btw.