Component score weighting for GMM based text-independent speaker verification

Liang Lu, Yuan Dong, Xianyu Zhao, Hao Yang, Jian Zhao, Haila Wang

Research output: Chapter in Book/Report/Conference proceedingConference contribution


GMM/UBM framework is wildly used in Automatic Speaker Verification (ASV), however, due to the insufficiency of the training data, both the hypothesized speaker and impostors are not well modeled, especially to some of the Gaussian component mixtures. Thus, the Gaussian mixtures in each GMM model have different discriminative capabilities, and the mismatch between testing and training data will also aggravate this situation. In this paper, we propose a novel approach, namely, Component Score Weighing (CSW), to reweight the Gaussian mixtures and highlight those which have high discriminative capability by post-processing the log-likelihood ratio (LLR). The original log-likelihood in GMM systems is assigned to each Gaussian component mixture, deriving two component score serials, which we called the dominant score serial and the residual score serial. A nonlinear score weighting function is then applied to reweigh those scores, respectively. Experiments on NIST 2006 SRE corpus show that, this approach achieves notable performance gains over our previous baseline system (about 12% relative improvement in minimum detection cost function (DCF) value).
Original languageEnglish
Title of host publicationOdyssey 2008: The Speaker and Language Recognition Workshop, Stellenbosch, South Africa, January 21-24, 2008
Number of pages1
Publication statusPublished - 2008
EventOdyssey 2008: The Speaker and Language Recognition Workshop - Stellenbosch, South Africa
Duration: 21 Jan 200824 Jan 2008


WorkshopOdyssey 2008: The Speaker and Language Recognition Workshop
CountrySouth Africa


Dive into the research topics of 'Component score weighting for GMM based text-independent speaker verification'. Together they form a unique fingerprint.

Cite this