A new penalty term for the BIC with respect to speaker diarization

Stafylakis, Themos; Tzimiropoulos, Georgios; Katsouros, Vassilis; Carayannis, George

Abstract

In this paper we examine a new penalty term for the Bayesian Information Criterion (BIC) that is suited to the problem of speaker diarization. Based on our previous approach of penalizing each cluster only with its effective sample size - an approach we called segmental - we propose a stricter penalty term. The criterion we derive retains the main property of the Segmental-BIC, i.e. it approximates the evidence of overall partitions of the data and simultaneously leads to a pairwise dissimilarity measure that is completely defined by the pair of clusters in question. The experimental results show significant improvement in diarization accuracy on the ESTER benchmark.

Citation

Stafylakis, T., Tzimiropoulos, G., Katsouros, V., & Carayannis, G. A new penalty term for the BIC with respect to speaker diarization. Presented at ICASSP 2010 - 2010 IEEE International Conference on Acoustics Speech and Signal Processing

• Conference Name: ICASSP 2010 - 2010 IEEE International Conference on Acoustics Speech and Signal Processing
• End Date: Mar 19, 2010
• Publication Date: Jan 1, 2010
• Deposit Date: Feb 1, 2016
• Publicly Available Date: Feb 1, 2016
• Peer Reviewed: Peer Reviewed
• Keywords: Bayes Methods, Speaker Recognition, Bayesian Information Criterion, Cluster Analysis, Speaker Diarization
• Public URL: https://nottingham-repository.worktribe.com/output/1013269
• Publisher URL: http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=5495076
• Additional Information: Published in: 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing: proceedings: March 14-19, 2010, Sheraton Dallas Hotel, Dallas, Texas, U.S.A. Piscataway, N.J. : IEEE, 2010. ISBN: 978-1-4244-4295-9. pp. 4978-4981, doi: 10.1109/ICASSP.2010.5495076 ©2010 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.