Title :
A multi-class MLLR kernel for SVM speaker recognition
Author :
Karam, Zahi N. ; Campbell, William M.
Author_Institution :
Lincoln Lab., MIT, Lexington, MA
fDate :
March 31 2008-April 4 2008
Abstract :
Speaker recognition using support vector machines (SVMs) with features derived from generative models has been shown to perform well. Typically, a universal background model (UBM) is adapted to each utterance yielding a set of features that are used in an SVM. We consider the case where the UBM is a Gaussian mixture model (GMM), and maximum likelihood linear regression (MLLR) adaptation is used to adapt the means of the UBM. Recent work has examined this setup for the case where a global MLLR transform is applied to all the mixture components of the QMM UBM. This work produced positive results that warrant examining this setup with multi-class MLLR adaptation, which groups the UBM mixture components into classes and applies a different transform to each class. This paper extends the MLLR/GMM framework to the multi- class case. Experiments on the NIST SRE 2006 corpus show that multi-class MLLR improves on global MLLR and that the proposed system´s performance is comparable with state of the art systems.
Keywords :
maximum likelihood estimation; regression analysis; speaker recognition; support vector machines; Gaussian mixture model; maximum likelihood linear regression; multiclass MLLR adaptation; speaker recognition; support vector machines; universal background model; Extraterrestrial measurements; Kernel; Laboratories; Maximum likelihood linear regression; NIST; Speaker recognition; Stacking; Support vector machine classification; Support vector machines; System performance; Adaptation; Kernel; MLLR; Speaker recognition; Support vector machine;
Conference_Titel :
Acoustics, Speech and Signal Processing, 2008. ICASSP 2008. IEEE International Conference on
Conference_Location :
Las Vegas, NV
Print_ISBN :
978-1-4244-1483-3
Electronic_ISBN :
1520-6149
DOI :
10.1109/ICASSP.2008.4518560