Research of speaker identification based on little training data

被引：0

作者：

Yang, YQ ^{[1
]}

Chen, W ^{[1
]}

Lu, YD ^{[1
]}

Gao, AG ^{[1
]}

机构：

[1] N China Elect Power Univ, Fac Control Sci & Engn, Baoding 071003, Peoples R China

来源：

PROCEEDINGS OF THE 2004 INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND CYBERNETICS, VOLS 1-7 | 2004年

关键词：

speaker identification; support vector machines; little training data; machine learning;

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Directing against little training data for speaker identification, this paper summarizes several current methods and analyses the existing problems. A new algorithm based on support vector machine is presented in the paper, and used to build a constrained text-independent speaker identification system. Experimental results indicate that the performance of test system is better than the system based on VQ, HMM or NN as comparison.

引用

页码：3755 / 3758

页数：4

共 6 条

[1] A tutorial on Support Vector Machines for pattern recognition [J].

Burges, CJC .

DATA MINING AND KNOWLEDGE DISCOVERY, 1998, 2 (02) :121-167

[2]

CAMPBELL C, 2000, P ESANN2000, P27

[3]

GISH H, 1994, IEEE SIGNAL PROC OCT, P18

[4]

Platt J., 1999, ADV KERNEL METHODS S

[5]

Vapnik V, 1999, NATURE STAT LEARNING

[6] A partitioned neural network approach for vowel classification using smoothed time frequency features [J].

Zahorian, SA ;

Nossair, ZB .

IEEE TRANSACTIONS ON SPEECH AND AUDIO PROCESSING, 1999, 7 (04) :414-425

← 1 →