Decision Making Based on Cohort Scores for Speaker Verification
This work addresses a specific problem in speaker verification systems for improving robustness to speech variations, representing an incremental advancement.
The paper tackles the sensitivity of single-score decision making in speaker verification to speech variations by proposing a decision approach based on multiple cohort scores, using a discriminative model like a DNN, which results in substantial performance improvement over the baseline system.
Decision making is an important component in a speaker verification system. For the conventional GMM-UBM architecture, the decision is usually conducted based on the log likelihood ratio of the test utterance against the GMM of the claimed speaker and the UBM. This single-score decision is simple but tends to be sensitive to the complex variations in speech signals (e.g. text content, channel, speaking style, etc.). In this paper, we propose a decision making approach based on multiple scores derived from a set of cohort GMMs (cohort scores). Importantly, these cohort scores are not simply averaged as in conventional cohort methods; instead, we employ a powerful discriminative model as the decision maker. Experimental results show that the proposed method delivers substantial performance improvement over the baseline system, especially when a deep neural network (DNN) is used as the decision maker, and the DNN input involves some statistical features derived from the cohort scores.