Presentation 2008-03-14
A Speaker Identification Method based on Trapezoid Fuzzy Similarity
Nguyen Thi Huong Lien, Yoshinori Arai, Fangyan Dong, Kaoru Hirota,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) A speaker identification method using trapezoid fuzzy membership function to calculate the similarity of voice features between a test people with his or her reference speaker model is proposed. The trapezoid fuzzy function is constructed based on the utterances recorded at three sessions of each user and a statistical data of a person recorded in a long period to absorb the variation. A set of acoustic voice features that can model a person is also introduced. This feature set is selected based on the ability of discriminate. The searching order is decided by the discriminate ability and feature calculation time.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Speaker Identification / acoustic feature / fuzzy membership / discriminate ability
Paper # SIS2007-82
Date of Issue

Conference Information
Committee SIS
Conference Date 2008/3/7(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Smart Info-Media Systems (SIS)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Speaker Identification Method based on Trapezoid Fuzzy Similarity
Sub Title (in English)
Keyword(1) Speaker Identification
Keyword(2) acoustic feature
Keyword(3) fuzzy membership
Keyword(4) discriminate ability
1st Author's Name Nguyen Thi Huong Lien
1st Author's Affiliation ()
2nd Author's Name Yoshinori Arai
2nd Author's Affiliation
3rd Author's Name Fangyan Dong
3rd Author's Affiliation
4th Author's Name Kaoru Hirota
4th Author's Affiliation
Date 2008-03-14
Paper # SIS2007-82
Volume (vol) vol.107
Number (no) 548
Page pp.pp.-
#Pages 4
Date of Issue