Boosting Speaker Recognition Performance with Compact Representations

This paper describes a speaker recognition system combination approach in which the compact forms of MAP adapted GMM supervectors are used to boost the performance of a highdimensional supervector-based system or a combination of multiple systems. The compact supervector representations are subjected to a diagonal transformation to emphasize those dimensions that describe significant speaker information and to deemphasize noisy dimensions. Scores obtained from these representations are then combined with the scores obtained from high-dimensional supervector representations. The transformation parameters and the combination weights are estimated by minimizing a discriminative training objective function that approximates a minimum detection cost function. We carried out experiments on two NIST 2008 Speaker Recognition Evaluation English telephony tasks to compare the proposed approach with direct score combination obtained from lowand highdimensional supervector representations. We have found that the proposed approach yields up to 18% relative gain.