Scatter Difference NAP for SVM Speaker Recognition
نویسندگان
چکیده
This paper presents Scatter Difference Nuisance Attribute Projection (SD-NAP) as an enhancement to NAP for SVM-based speaker verification. While standard NAP may inadvertently remove desirable speaker variability, SD-NAP explicitly de-emphasises this variability by incorporating a weighted version of the between-class scatter into the NAP optimisation criterion. Experimental evaluation of SD-NAP with a variety of SVM systems on the 2006 and 2008 NIST SRE corpora demonstrate that SD-NAP provides improved verification performance over standard NAP in most cases, particularly at the EER operating point.
منابع مشابه
Discriminant NAP for SVM speaker recognition
Nuisance Attribute Projection (NAP) provides an effective method of removing the unwanted session variability in a Support Vector Machine (SVM) based speaker recognition system by removing the principal components of this variability. There is no guarantee with the methods proposed, however, that desired speaker variability is retained. This paper investigates the possibility of training NAP di...
متن کاملVariability compensated support vector machines applied to speaker verification
Speaker verification using SVMs has proven successful, specifically using the GSV Kernel [1] with nuisance attribute projection (NAP) [2]. Also, the recent popularity and success of joint factor analysis [3] has led to promising attempts to use speaker factors directly as SVM features [4]. NAP projection and the use of speaker factors with SVMs are methods of handling variability in SVM speaker...
متن کاملLinear and non linear kernel GMM supervector machines for speaker verification
This paper presents a comparison between Support Vector Machines (SVM) speaker verification systems based on linear and non linear kernels defined in GMM supervector space. We describe how these kernel functions are related and we show how the nuisance attribute projection (NAP) technique can be used with both of these kernels to deal with the session variability problem. We demonstrate the imp...
متن کاملALIZE/spkdet: a state-of-the-art open source software for speaker recognition
This paper presents the ALIZE/SpkDet open source software packages for text independent speaker recognition. This software is based on the well-known UBM/GMM approach. It includes also the latest speaker recognition developments such as Latent Factor Analysis (LFA) and unsupervised adaptation. Discriminant classifiers such as SVM supervectors are also provided, linked with the Nuisance Attribut...
متن کاملThe NIST SRE summed channel speaker recognition system
This paper presents an improved speaker recognition system for the summed channel evaluation tasks in the 2008 NIST SRE (SRE08) with multiple summed-channel excerpts for speaker training and one summed-channel excerpt for testing. The system includes three main modules in which a hybrid speaker purification and clustering algorithm is adopted to segregate the summed-channel speech, a common spe...
متن کامل