Compositional Embedding Models for Speaker Identification and Diarization with Simultaneous Speech From 2+ Speakers.

Zeqian Li, Jacob Whitehill

Browse the full ICASSP paper archive.