Parameterization of the score threshold for a text-dependent adaptive speaker verification system

In this work we present a computationally efficient strategy for setting a priori thresholds in an adaptive speaker verification system. Our motivations are two-fold: one is to eliminate the externally pre-set overall system thresholds and replace them with automatically-set internal thresholds conditioned by a target FA rate and calculated at runtime, and the other is to counter the verification score shifts resulting from online adaptation. Our approach entails calculating the trajectory of the score threshold as a function of 1) length of the password, 2) target FA, and 3) the number of training frames in the speaker model. The solution is successful at both achieving target FA rates and keeping the FA rate constant during online adaptation. Furthermore, it is algorithmically simple and requires negligible computational resources. The threshold function is calibrated on a Japanese database and experimental results are presented on 12 databases in four different languages.