Towards the use of full covariance models for missing data speaker recognition.

Marco Khne, Daniel Pullella, Roberto Togneri, Sven Nordholm

Browse the full ICASSP paper archive.