Computationally Efficient and Robust BIC-Based Speaker Segmentation

File Description SizeFormat 
IEEE_TRANS_ASLP_2008_Margarita_Kotti.pdfAccepted version334.93 kBAdobe PDFView/Open
Title: Computationally Efficient and Robust BIC-Based Speaker Segmentation
Authors: Kotti, M
Benetos, E
Kotropoulos, C
Item Type: Journal Article
Abstract: An algorithm for automatic speaker segmentation based on the Bayesian information criterion (BIC) is presented. BIC tests are not performed for every window shift, as previously, but when a speaker change is most probable to occur. This is done by estimating the next probable change point thanks to a model of utterance durations. It is found that the inverse Gaussian fits best the distribution of utterance durations. As a result, less BIC tests are needed, making the proposed system less computationally demanding in time and memory, and considerably more efficient with respect to missed speaker change points. A feature selection algorithm based on branch and bound search strategy is applied in order to identify the most efficient features for speaker segmenta-tion. Furthermore, a new theoretical formulation of BIC is derived by applying centering and simultaneous diagonalization. This formulation is considerably more computationally efficient than the standard BIC, when the covariance matrices are estimated by other estimators than the usual maximum-likelihood ones. Two commonly used pairs of figures of merit are employed and their relationship is established. Computational efficiency is achieved through the speaker utterance modeling, whereas robustness is achieved by feature selection and application of BIC tests at appropriately selected time instants. Experimental results indicate that the proposed modifications yield a superior performance compared to existing approaches. © 2008 IEEE.
Issue Date: 1-Dec-2008
ISSN: 1558-7916
Publisher: IEEE
Start Page: 920
End Page: 933
Journal / Book Title: IEEE Transactions on Audio, Speech, and Language Processing
Volume: 16
Issue: 5
Copyright Statement: © 2008 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.
Appears in Collections:Division of Surgery

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Creative Commons