Time delay deep neural network-based universal background models for speaker recognition.

David Snyder, Daniel Garcia-Romero, Daniel Povey

Browse the full ASRU paper archive.