On performance improvement of a speaker verification system using vector quantization, cohorts and hybrid cohort-world models

• Published in: International journal of speech technology Vol. 5; no. 3; pp. 247 - 257
• Main Authors: Burileanu, Corneliu, Moraru, Daniel, Bojan, Luigi, Puchiu, Mihai, Stan, Alina
• Format: Journal Article
• Language: English
• Published: 2002
• ISSN: 1381-2416
• DOI: 10.1023/A:1020244924468

This paper presents the use of distance normalization techniques in order to improve speaker verification system performance. These techniques provide a dynamic threshold that compensates for the trial-to-trial variations & replaces the fixed threshold used in the classical speaker verification approach. Two methods are described: the cohort model normalization & a new & original hybrid cohort-world model normalization. These methods are compared from the point of view of storage space requirements & computational effort. Two algorithms are proposed: one uses existing user models, & the other creates new models. The algorithms were evaluated using the YOHO database & a proprietary database. The results showed that using these methods, the errors of false rejection are significantly reduced for a constant false acceptance error, when the cohort size is increasing. The algorithms also involve fewer computational resources than other algorithms, making them more suitable for commercial application. 13 Figures, 16 References. Adapted from the source document