Speaker identification through use of features selected using genetic algorithm

被引：8

作者：

Haydar, A ^{[1
]}

Demirekler, M

Yurtseven, MK

机构：

[1] Eastern Mediterranean Univ, Elect & Elect Engn Dept, Gazi Magosa, Cyprus

[2] Middle E Tech Univ, Dept Elect & Elect Engn, TR-06531 Ankara, Turkey

来源：

ELECTRONICS LETTERS | 1998年 / 34卷 / 01期

关键词：

D O I：

10.1049/el:19980126

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

The authors introduce the use of a genetic algorithm in the reduction of a 24 parameter (12 LPC derived cepstral coefficients +12 Delta-cepstral coefficients) set to a five, six, seven, eight or ten parameter set, for each speaker in text-independent speaker identification. The experimental results show that there is similar to 5% increase in the recognition rate when the reduced set of parameters is used.

引用

页码：39 / 40

页数：2

共 5 条

[1]

BIDSARIA HB, 1987, PATTERN RECOGN, V20, P365

[2] ON TEXT INDEPENDENT SPEAKER IDENTIFICATION USING A QUADRATIC CLASSIFIER WITH OPTIMAL FEATURES [J].

COHEN, A ;

FROIND, I .

SPEECH COMMUNICATION, 1989, 8 (01) :35-44

[3]

HAYDAR A, 1995, 5 INT C ADV COMM CON, P56

[4] TOWARD A FUNDAMENTAL THEORY OF OPTIMAL FEATURE-SELECTION .1. [J].

MORGERA, SD ;

DATTA, L .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1984, 6 (05) :601-616

[5]

RABINER LR, 1987, DIGITAL PROCESSING S, P96

← 1 →