Presentation 2001/12/13
A study of speaker segmentation of dialogue speech with speech overlapped section
Masahumi KOBAYASHI, Tatsuya KITAMURA, Shigeyoshi KITAZAWA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, speaker segmentation of dialogue speech was studied to develop a support system for dialogue analysis. The dialogue speech addressed in this study was spoken by two speakers, and had speech overlapped section. To train HMM for the overlapped section, two different speech databases were constructed: (1)same words spoken by the two speakers were added;(2)pairs of different words spoken by the two speakers were added. These two training approaches were compared using dialogue speech data. The experimental result showed the latter have better performance. In addition, removing redundant arcs on the HMM network was improved the performance.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) multispeaker / speaker recognition / speaker segmentation / dialogue speech
Paper # NLC2001-54,SP2001-89
Date of Issue

Conference Information
Committee NLC
Conference Date 2001/12/13(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 Natural Language Understanding and Models of Communication (NLC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A study of speaker segmentation of dialogue speech with speech overlapped section
Sub Title (in English)
Keyword(1) multispeaker
Keyword(2) speaker recognition
Keyword(3) speaker segmentation
Keyword(4) dialogue speech
1st Author's Name Masahumi KOBAYASHI
1st Author's Affiliation Graduate School of Informatics, Shizuoka University()
2nd Author's Name Tatsuya KITAMURA
2nd Author's Affiliation Faculty of Information, Shizuoka University
3rd Author's Name Shigeyoshi KITAZAWA
3rd Author's Affiliation Faculty of Information, Shizuoka University
Date 2001/12/13
Paper # NLC2001-54,SP2001-89
Volume (vol) vol.101
Number (no) 520
Page pp.pp.-
#Pages 6
Date of Issue