is also described and discussed. Finally, representative performance benchmarks and system behavior experiments on NIST SRE corpora are presented. © 2000 Academic Press Key Words: speaker recognition; Gaussian mixture models; likelihood ratio detector; universal background model; handset normalization

Uncertainty propagation for noise robust speaker

BUT submitted three systems to NIST SRE 2008 eval-uations, only to the short2-short3 condition. The pri-mary system is a fusion of three sub-systems: 2 based on MFCC and factor analysis and one making use of SVM

-gram probabilities. Multiple vectors are concatenated to form a composite vector. Finally the Support Vector Machine (SVM) is used as the back-end classifier of the composite vectors. We use the 2002 NIST Speaker Recognition Evaluation (SRE) corpus for training GMM tokenizers and background modeling, and evaluate

and often overlapping, and the microphones are inconspicuously placed. To support the development of meeting recognition technologies by both the speech recognition and video extraction research communities, NIST is providing a development and evaluation infrastructure including: a multi-media corpus

between the method of SVM nuisance attribute projection (NAP) and the recent results in latent factor analysis. Experiments on a NIST SRE 2005 corpus demonstrate the effectiveness of the new technique. 1.

Fusion of the base classifiers is seen as a way to achieve high performance in state-of-the-art speaker verification systems. Typically, we are looking for base classifiers that would be com-plementary. We might also be interested in reinforcing good base classifiers by including others that are similar to them. In any case, the final ensemble size is typically small and has to be formed based on some rules of thumb. We are interested to find out a subset of classifiers that has a good generalization perfor-mance. We approach the problem from sparse learning point of view. We assume that the true, but unknown, fusion weights are sparse. As a practical solution, we regularize weighted logistic regression loss function by elastic-net and LASSO constraints. However, all regularization methods have an additional param-eter that controls the amount of regularization employed. This needs to be separately tuned. In this work, we use variational Bayes approach to automatically obtain sparse solutions without additional cross-validation. Variational Bayes method improves the baseline method in 3 out of 4 sub-conditions. Index Terms: logistic regression, regularization, compressed sensing, linear fusion, speaker verification

For the participation in SRE 2012 the Radboud University Nijmegen (RUN) team was part of a larger effort “I4U, ” to which we contributed development test segment lists and score matrices used for fusion in the I4U submission. This system description is about the RUN system, for which the scores

approach, a gradient descent algorithm is proposed to find out projection over input variables. Experimental results on the 2006 NIST SRE corpus show that both kinds of NAP can reduce unwanted variability in nonlinear kernels to improve verification performance; and NAP performed in expanded feature space

information in speaker characterization, the novel feature set has been tested in the 2010 NIST Speaker Recognition Evaluation (NIST SRE10). The proposed system uses glottal estimate parameter templates and classical cepstral information to build a model for each speaker involved in the recognition process