Baihan Lin(University of Washington), Xinxin Zhang(University of Washington)
Abstract:
We proposed a novel AI framework to conduct real-time multi-speaker recognition without any prior registration or pretraining by learning the speaker identification on the fly. We considered the practical problem of online learning with episodically revealed rewards and introduced a solution based on semi-supervised and self-supervised learning methods in a web based application at https://www.baihan.nyc/viz/VoiceID/.