PL-EESR: Perceptual Loss Based End-to-End Robust Speaker Representation Extraction.

Yi Ma, Kong Aik Lee, Ville Hautamki, Haizhou Li

Browse the full ASRU paper archive.