1 paper
Joon Son Chung, Jaesung Huh, Seongkyu Mun +7
The objective of this paper is 'open-set' speaker recognition of unseen speakers, where ideal embeddings should be able to condense information into a compact utterance-level repre…