PLDA-based mean shift speakers' short segments clustering. (September 2017)

• Record Type: Journal Article
• Authors: Salmun, Itay; Shapiro, Ilya; Opher, Irit; Lapidot, Itshak
• Is Part Of: Computer speech & language. Volume 45(2017)
• Page Start: 411
• Page End: 436
• Publication Date: 2017-09
• Subjects: Speaker clustering -- Mean shift clustering -- Probabilistic linear discriminant analysis -- Two-covariance model -- K-nearest neighbors -- I-vectors -- Short segments; Speech processing systems -- Periodicals; Automatic speech recognition -- Periodicals; Computers -- Periodicals; Linguistics -- Periodicals; Speech-Language Pathology -- Periodicals; Traitement automatique de la parole -- Périodiques; Reconnaissance automatique de la parole -- Périodiques; Automatic speech recognition; Speech processing systems; Electronic journals; Periodicals; 006.454
• DOI: 10.1016/j.csl.2017.04.006
• Languages: English
• ISSNs: 0885-2308

Abstract

Highlights: Use of PLDA as a criterion to choose the best i-vectors for the new mean. Replacing the constant threshold for as the i-vector's neighborhood by the kNN which dramatically increased the stability of the system. Abstract: This paper extends upon a previous work using Mean Shift algorithm to perform speaker clustering on i-vectors generated from short speech segments. In this paper we examine the effectiveness of probabilistic linear discriminant analysis (PLDA) scoring as the metric of the mean shift clustering algorithm in the presence of different numbers of speakers. Our proposed method, combined with k-nearest neighbors (kNN) for bandwidth estimation, yields better and more robust results in comparison to the cosine similarity with fixed neighborhood bandwidth for clustering segments of large numbers of speakers. In the case of 30 speakers, we achieved significant improvement in cluster and speaker purity with the PLDA-based mean shift algorithm compared to the cosine-based baseline system.