A session-GMM generative model using test utterance Gaussian Mixture Modeling for speaker verification
Hagai Aronowitz, David Burshtein and Amihood Amir
In: ICASSP 2005, 19-23 Mar 2005, Philadelphia, PA, USA.
Test-utterance parameterization (TUP) using Gaussian Mixture Models (GMMs) has recently shown to be beneficial for speaker indexing due to its computational efficiency and identical accuracy compared to classic GMM-based recognizers. In this paper we show that TUP can also lead to more accurate speaker recognition. On the NIST-2004 evaluation corpus, recognition error rate was reduced by 8% compared to the classic GMM-based algorithm. Furthermore, we introduce a novel generative statistical model for generation of test utterances by speakers. This model is incorporated naturally into the TUP framework and improves speaker recognition accuracy. On the NIST-2004 evaluation corpus, recognition error rate was reduced by 15% compared to the classic GMM-based algorithm.