Compensation of Nuisance Factors for Speaker and Language Recognition

Abstract

The variability of the channel and environment is one of the most important factors affecting the performance of text-independent speaker verification systems. The best techniques for channel compensation are model based. Most of them have been proposed for Gaussian Mixture Models, while in the feature domain blind channel compensation is usually performed. The aim of this work is to explore techniques that allow more accurate intersession compensation in the feature domain. Compensating the features rather than the models has the advantage that the transformed parameters can be used with models of a different nature and complexity, and for different tasks. In this paper, we evaluate the effects of the compensation of the intersession variability obtained by means of the channel factors approach. In particular, we compare channel variability modeling in the usual Gaussian Mixture model domain, and our proposed feature domain compensation technique. We show that the two approaches lead to similar results on the NIST 2005 Speaker Recognition Evaluation data with a reduced computation cost. We report also the results of a system, based on the intersession compensation technique in the feature space that was among the best participants in the NIST 2006 Speaker Recognition Evaluation. Moreover, we show how we obtained significant performance improvement in language recognition by estimating and compensating, in the feature domain, the distortions due to inter-speaker variability within the same language.

A Study of Interspeaker Variability in Speaker Verification

“…It turns out that, in the case of a diagonal factor analysis model, using speaker-dependent GMM's does indeed produce better results. For example, on the English language trials in the core condition, a diagonal model with 100 channel factors produces an EER of 2.8% for male speakers, which is similar to the results presented in [30], [5], [31] (but not as good as the result in the first line of Table V).…”

“…The results we have obtained using speaker factors are clearly much better than those obtained using d alone, but the reader may have noticed that the figures presented in the fourth rows of Tables I and II are not quite as good as the best results that have been reported with comparable standalone GMM/UBM systems as in [30], [5], [31]. These systems are comparable because they use relevance MAP for speaker enrollment and channel factors to compensate for intersession variability.…”

“…Although most authors (e.g. [4], [5]) use the term factor analysis to refer to the channel model (3) alone, we have always used this term in a broader sense which includes the speaker model (2) as well. (Where it is necessary to make this distinction explicitly we speak of joint factor analysis.)…”

Audio segmentation-by-classification approach based on factor analysis in broadcast news domain

“…Most of the approaches based on FA for language recognition are implemented with a single U matrix because the segments are well-delimited (typically in separated files) and the nature of the withinclass variability is similar for all the languages as it can be seen in [36,[46][47][48]. In [49], a segmentation system was proposed with five class location vectors (one vector per class) and a single compensation matrix U for all the classes.…”

Double Gauss Based Unsupervised Score Normalization in Speaker Verification

“…The factor analysis [6,7,8] has been found to be very effective in reducing the channel bias in the Gaussian mixture models and universal background models (GMM-UBM) system [6]. In this paper, we use factor analysis to reduce channel bias.…”