CONF vijayasenan:ASRU:2007/IDIAP AGGLOMERATIVE INFORMATION BOTTLENECK FOR SPEAKER DIARIZATION OF MEETINGS DATA Vijayasenan, Deepu Valente, Fabio Bourlard, Hervé EXTERNAL http://publications.idiap.ch/attachments/papers/2007/vijayasenan-ASRU-2007.pdf PUBLIC http://publications.idiap.ch/index.php/publications/showcite/vijayasenan:rr07-31 Related documents IEEE Automatic Speech Recognition and Understanding Workshop 2007 IDIAP-RR 07-31 In this paper, we investigate the use of agglomerative Information Bottleneck (aIB) clustering for the speaker diarization task of meetings data. In contrary to the state-of-the-art diarization systems that models individual speakers with Gaussian Mixture Models, the proposed algorithm is completely non parametric . Both clustering and model selection issues of non-parametric models are addressed in this work. The proposed algorithm is evaluated on meeting data on the RT06 evaluation data set. The system is able to achieve Diarization Error Rates comparable to state-of-the-art systems at a much lower computational complexity. REPORT vijayasenan:rr07-31/IDIAP AGGLOMERATIVE INFORMATION BOTTLENECK FOR SPEAKER DIARIZATION OF MEETINGS DATA Vijayasenan, Deepu Valente, Fabio Bourlard, Hervé EXTERNAL http://publications.idiap.ch/attachments/reports/2007/vijayasenan-idiap-rr-07-31.pdf PUBLIC Idiap-RR-31-2007 2007 IDIAP In this paper, we investigate the use of agglomerative Information Bottleneck (aIB) clustering for the speaker diarization task of meetings data. In contrary to the state-of-the-art diarization systems that models individual speakers with Gaussian Mixture Models, the proposed algorithm is completely non parametric . Both clustering and model selection issues of non-parametric models are addressed in this work. The proposed algorithm is evaluated on meeting data on the RT06 evaluation data set. The system is able to achieve Diarization Error Rates comparable to state-of-the-art systems at a much lower computational complexity.