Minimum classification error/eigenvoices training for speaker identification.

Fabio Valente, Christian Wellekens

Browse the full ICASSP paper archive.