This pairs you with one random stranger over live voice, and you each try to guess where the other is from by their accent. The audio never touches my server. The voice goes peer to peer over WebRTC, and the server only does the matchmaking and passes the signaling messages back and forth.
I'd really like feedback, especially on the connection edge cases and anything that falls over. Thanks for taking a look.
rolph•1h ago
you can say with confidence,a person has deepexposure to a set of dialect and affect, however there is reduced confidence when using voice character to infer country of origin, or current location.
townload•1h ago