[tootie]

Believe it or not, overlayed closed captions was one the first things I came up with. It's also not that hard to do with commodity voice and face recognition. We did a POC just on a 2d phone screen in like a week. Trying to capture multiple people speaking at once is way beyond the capability of any retail headset and would require an elaborate 3d microphone array and noise filtering to pinpoint where a voice is coming from. Ours worked pretty well sitting across a table, but would struggle mightily trying to hear something across any distance in a noisy room.