A. Stolcke, L. Ferrer and S. Kajarekar, “Improvements in MLLR-Transform-based Speaker Recognition,” 2006 IEEE Odyssey – The Speaker and Language Recognition Workshop, 2006, pp. 1-6, doi: 10.1109/ODYSSEY.2006.248089.
Abstract
We previously proposed the use of MLLR transforms derived from a speech recognition system as speaker features in a speaker verification system. In this paper we report recent improvements to this approach. First, we noticed a fundamental problem in our previous implementation that stemmed from a mismatch between male and female recognition models, and the model transforms they produce. Although it affects only a small percentage of verification trials (those in which the gender detector commits errors), this mismatch has a large effect on average system performance. We solve this problem by consistently using only one recognition model (either male or female) in computing speaker adaptation transforms regardless of estimated speaker gender. A further accuracy boost is obtained by combining feature vectors derived from male and female vectors into one larger feature vector. Using 1-conversation-side
training, the final system has about 27% lower decision cost than a state-of-the-art cepstral GMM speaker system, and 53% lower decision cost when trained on 8 conversation sides per speaker.