Multiple Speakers from a single "feed"

I will typically host a session where someone on a video call will share the call with another person. Think of a husband and wife, Owner and Co-Owner, etc.; the system will see them as a single person and voice. It would be nice if the system could detect the different “voices” and apply their unique persona. I’d happily tag person one as “Tom” and person two as “Vicki” if that made it easier.