Codebook design method for speaker identification based on Genetic Algorithm

Author

Islam, Md Rabiul ; Rahman, Md Fayzur ; Amin, Md Ruhul ; Zaman, Md Shahid Uz

Author_Institution

Dept. of Comput. Sci. & Eng., Rajshahi Univ. of Eng. & Technol. (RUET), Rajshahi

fYear

2008

fDate

24-27 Dec. 2008

Firstpage

116

Lastpage

120

Abstract

In this paper, a novel method of designing a codebook for speaker identification purpose utilizing genetic algorithm has been proposed. Speech features have been extracted using standard speech parameterization method such as LPC, LPCC, RCC, MFCC, DeltaMFCC and DeltaDeltaMFCC. For each of these techniques, the performance of the proposed system has been compared. In this codebook design method, genetic algorithm has the capability of getting global optimal result and hence improves the quality of the codebook. Experimental results show the superiority (i.e. 92% accuracy has been achieved) of this proposed codebook design method.

Keywords

feature extraction; genetic algorithms; speaker recognition; codebook design method; genetic algorithm; speaker identification; speech feature extraction; speech parameterization method; Acoustic distortion; Algorithm design and analysis; Decoding; Design engineering; Design methodology; Genetic algorithms; Genetic engineering; Loudspeakers; Mel frequency cepstral coefficient; Speech; Automatic speaker identification; codebook design; genetic algorithm; speech parameterization; speech pre-processing;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer and Information Technology, 2008. ICCIT 2008. 11th International Conference on

Conference_Location

Khulna

Print_ISBN

978-1-4244-2135-0

Electronic_ISBN

978-1-4244-2136-7

Type

conf

DOI

10.1109/ICCITECHN.2008.4803002

Filename

4803002