Speaker age estimation using Hidden Markov Model weight supervectors

Author

Bahari, Mohamad Hasan ; Van hamme, Hugo

Author_Institution

Dept. of Electr. Eng. (ESAT), KU Leuven, Leuven, Belgium

fYear

2012

fDate

2-5 July 2012

Firstpage

517

Lastpage

521

Abstract

This paper proposes a new approach for speaker age estimation. In this method, speakers are modeled by their corresponding Hidden Markov Model (HMM) weight supervectors. Then, Weighted Supervised Non-Negative Matrix Factorization (WSNMF) is applied to reduce the dimension of the input space. Finally, a Least Squares Support Vector Regressor (LS-SVR) is employed to estimate the age of speakers using the obtained low-dimensional vectors. Evaluation results on a corpus of read and spontaneous speech in Dutch confirms the effectiveness of the proposed scheme.

Keywords

age issues; hidden Markov models; least squares approximations; matrix decomposition; regression analysis; speaker recognition; vectors; Dutch; HMM weight supervectors; LS-SVR; WSNMF; hidden Markov model weight supervectors; input space dimension reduction; least squares support vector regressor; low-dimensional vectors; speaker age estimation; weighted supervised nonnegative matrix factorization; Acoustics; Estimation; Hidden Markov models; Speech; Support vector machines; Training; Wireless sensor networks; Least Squares Support Vector Regressor; speaker age estimation; weighted supervised non-negative matrix factorization;

fLanguage

English

Publisher

ieee

Conference_Titel

Information Science, Signal Processing and their Applications (ISSPA), 2012 11th International Conference on

Conference_Location

Montreal, QC

Print_ISBN

978-1-4673-0381-1

Electronic_ISBN

978-1-4673-0380-4

Type

conf

DOI

10.1109/ISSPA.2012.6310606

Filename

6310606