CONF Vijayasenan_INTERSPEEH2009_2009/IDIAP KL Realignment for Speaker Diarization with Multiple Feature Streams Vijayasenan, Deepu Valente, Fabio Bourlard, Hervé http://publications.idiap.ch/index.php/publications/showcite/Vijayasenan_Idiap-RR-24-2010 Related documents 10th Annual Conference of the International Speech Communication Association 2009 This paper aims at investigating the use of Kullback-Leibler (KL) divergence based realignment with application to speaker diarization. The use of KL divergence based realignment operates directly on the speaker posterior distribution estimates and is compared with traditional realignment performed using HMM/GMM system. We hypothesize that using posterior estimates to re-align speaker boundaries is more robust than gaussian mixture models in case of multiple feature streams with different statistical properties. Experiments are run on the NIST RT06 data. These experiments reveal that in case of conventional MFCC features the two approaches yields the same performance while the KL based system outperforms the HMM/GMM re-alignment in case of combination of multiple feature streams (MFCC and TDOA). REPORT Vijayasenan_Idiap-RR-24-2010/IDIAP KL Realignment for Speaker Diarization with Multiple Feature Streams Vijayasenan, Deepu Valente, Fabio Bourlard, Hervé EXTERNAL http://publications.idiap.ch/attachments/reports/2009/Vijayasenan_Idiap-RR-24-2010.pdf PUBLIC Idiap-RR-24-2010 2010 Idiap July 2010 This paper aims at investigating the use of Kullback-Leibler (KL) divergence based realignment with application to speaker diarization. The use of KL divergence based realignment operates directly on the speaker posterior distribution estimates and is compared with traditional realignment performed using HMM/GMM system. We hypothesize that using posterior estimates to re-align speaker boundaries is more robust than gaussian mixture models in case of multiple feature streams with different statistical properties. Experiments are run on the NIST RT06 data. These experiments reveal that in case of conventional MFCC features the two approaches yields the same performance while the KL based system outperforms the HMM/GMM re-alignment in case of combination of multiple feature streams (MFCC and TDOA).