Support vector regression for speaker verification

Proceedings of Interspeech 2007, Antwerp (Belgium)This paper explores Support Vector Regression (SVR) as an alternative to the widely-used Support Vector Classification (SVC) in GLDS (Generalized Linear Discriminative Sequence)-based speaker verification. SVR allows the use of a ε-insensitive loss function which presents many advantages. First, the optimization of the ε parameter adapts the system to the variability of the features extracted from the speech. Second, the approach is robust to outliers when training the speaker models. Finally, SVR training is related to the optimization of the probability of the speaker model given the data. Results are presented using the NIST SRE 2006 protocol, showing that SVR-GLDS yields a relative improvement of 31% in EER compared to SVC-GLDS.This work was partially funded by the Spanish Ministry of Education under project TEC2006-13170-C02-01