The IIR submission to CSLP 2006 speaker recognition evaluation

Kong Aik Lee, Hanwu Sun, Rong Tong, Bin Ma, Minghui Dong, Changhuai You, Donglai Zhu, Chin Wei Eugene Koh, Lei Wang, Tomi Kinnunen, Eng Siong Chng, Haizhou Li

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

Abstract

This paper describes the design and implementation of a practical automatic speaker recognition system for the CSLP speaker recognition evaluation (SRE). The speaker recognition system is built upon four subsystems using speaker information from acoustic spectral features. In addition to the conventional spectral features, a novel temporal discrete cosine transform (TDCT) feature is introduced in order to capture long-term speech dynamic. The speaker information is modeled using two complementary speaker modeling techniques, namely, Gaussian mixture model (GMM) and support vector machine (SVM). The resulting subsystems are then integrated at the score level through a multilayer perceptron (MLP) neural network. Evaluation results confirm that the feature selection, classifier design, and fusion strategy are successful, giving rise to an effective speaker recognition system.

Original languageEnglish
Title of host publicationChinese Spoken Language Processing - 5th International Symposium, ISCSLP 2006, Proceedings
Pages494-505
Number of pages12
DOIs
Publication statusPublished - Dec 2006
Externally publishedYes
Event5th International Symposium on Chinese Spoken Language Processing, ISCSLP 2006 - Singapore, Singapore
Duration: 13 Dec 200616 Dec 2006

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume4274 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference5th International Symposium on Chinese Spoken Language Processing, ISCSLP 2006
Country/TerritorySingapore
CitySingapore
Period13/12/0616/12/06

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'The IIR submission to CSLP 2006 speaker recognition evaluation'. Together they form a unique fingerprint.

Cite this