CONF Motlicek_ICASSP_2015/IDIAP EMPLOYMENT OF SUBSPACE GAUSSIAN MIXTURE MODELS IN SPEAKER RECOGNITION Motlicek, Petr Dey, Subhadeep Madikeri, Srikanth Burget, Lukas EXTERNAL https://publications.idiap.ch/attachments/papers/2015/Motlicek_ICASSP_2015.pdf PUBLIC https://publications.idiap.ch/index.php/publications/showcite/Motlicek_Idiap-RR-16-2015 Related documents IEEE - 2015 IEEE International Conference on Acoustics, Speech, and Signal Processing Brisbane, Australia 2015 4445-4449 978-1-4673-6996-1 http://icassp2015.org/ URL This paper presents Subspace Gaussian Mixture Model (SGMM) approach employed as a probabilistic generative model to estimate speaker vector representations to be subsequently used in the speaker verification task. SGMMs have already been shown to significantly outperform traditional HMM/GMMs in Automatic Speech Recognition (ASR) applications. An extension to the basic SGMM framework allows to robustly estimate low-dimensional speaker vectors and exploit them for speaker adaptation. We propose a speaker verification framework based on low-dimensional speaker vectors estimated using SGMMs, trained in ASR manner using manual transcriptions. To test the robustness of the system, we evaluate the proposed approach with respect to the state-of-the-art i-vector extractor on the NIST SRE 2010 evaluation set and on four different length-utterance conditions: 3sec-10sec, 10 sec-30 sec, 30 sec-60 sec and full (untruncated) utterances. Experimental results reveal that while i-vector system performs better on truncated 3sec to 10sec and 10 sec to 30 sec utterances, noticeable improvements are observed with SGMMs especially on full length-utterance durations. Eventually, the proposed SGMM approach exhibits complementary properties and can thus be efficiently fused with i-vector based speaker verification system. REPORT Motlicek_Idiap-RR-16-2015/IDIAP EMPLOYMENT OF SUBSPACE GAUSSIAN MIXTURE MODELS IN SPEAKER RECOGNITION Motlicek, Petr Dey, Subhadeep Madikeri, Srikanth Burget, Lukas Automatic Speech Recognition i-vectors speaker recognition subspace Gaussian mixture models EXTERNAL https://publications.idiap.ch/attachments/reports/2015/Motlicek_Idiap-RR-16-2015.pdf PUBLIC Idiap-RR-16-2015 2015 Idiap Rue Marconi 19, Martigny June 2015 This paper presents Subspace Gaussian Mixture Model (SGMM) approach employed as a probabilistic generative model to estimate speaker vector representations to be subsequently used in the speaker verification task. SGMMs have already been shown to significantly outperform traditional HMM/GMMs in Automatic Speech Recognition (ASR) applications. An extension to the basic SGMM framework allows to robustly estimate low-dimensional speaker vectors and exploit them for speaker adaptation. We propose a speaker verification framework based on low-dimensional speaker vectors estimated using SGMMs, trained in ASR manner using manual transcriptions. To test the robustness of the system, we evaluate the proposed approach with respect to the state-of-the-art i-vector extractor on the NIST SRE 2010 evaluation set and on four different length-utterance conditions: 3sec-10sec, 10 sec-30 sec, 30 sec-60 sec and full (untruncated) utterances. Experimental results reveal that while i-vector system performs better on truncated 3sec to 10sec and 10 sec to 30 sec utterances, noticeable improvements are observed with SGMMs especially on full length-utterance durations. Eventually, the proposed SGMM approach exhibits complementary properties and can thus be efficiently fused with i-vector based speaker verification system.