Speaker identification with vector quantization and k-harmonic means

Author

Yazici, Mustafa ; Ulutas, Mustafa

Author_Institution

Bilgisayar Muhendisligi Bolumu, Karadeniz Teknik Univ., Trabzon, Turkey

fYear

2014

fDate

23-25 April 2014

Firstpage

2134

Lastpage

2137

Abstract

A new method is proposed in this study to identify speakers in a relatively short time without decreasing success ratio. The method first extracts MFCC (Mel Frequency Cepstrum Coefficients) features. Then k-harmonic means is used to cluster samples before classification is performed by the nearest neighbor method. International HYKE database is used to test the performance of the proposed method in terms of success ratio and runtime, and compare with both MFCC+k-means and MFCC+LBG methods. Preliminary results show that the proposed method usually gives the best performance.

Keywords

audio databases; feature extraction; pattern clustering; signal classification; speaker recognition; vector quantisation; MFCC feature extraction; MFCC-LBG methods; MFCC-k-means; Mel frequency cepstrum coefficient features; classification; international HYKE database; k-harmonic means; nearest neighbor method; sample clustering; speaker identification; vector quantization; Biometrics (access control); Conferences; Feature extraction; Mel frequency cepstral coefficient; Speaker recognition; Vector quantization; k-harmonic means; speaker identification; vector quantization;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing and Communications Applications Conference (SIU), 2014 22nd

Conference_Location

Trabzon

Type

conf

DOI

10.1109/SIU.2014.6830684

Filename

6830684