Discriminative Feedback Adaptation for GMM-UBM Speaker Verification
Yi-Hsiang Chao, Wei-Ho Tsai, Hsin‐Min Wang · 2008
The GMM-UBM system is the current state-of-the-art approach for text-independent speaker verification. The advantage of the approach is that both target speaker model and impostor model (UBM) have generalization ability to handle "unseen" acoustic patterns. However, since GMM-UBM uses a common anti-model, namely UBM, for all target speakers, it tends to be weak in rejecting impostors' voices that are similar to the target speaker's voice. To overcome this limitation, we propose a discriminative feedback adaptation (DFA) framework that reinforces the discriminability between the target speaker model and the anti- model, while preserves the generalization ability of the GMM-UBM approach. This is done by adapting the UBM to a target-speaker- dependent anti-model based on a minimum verification squared- error criterion, rather than estimating from scratch by applying the conventional discriminative training schemes. The results of experiments conducted on the NTST2001-SRE database show that DFA substantially improves the performance of the conventional GMM-UBM approach.