Presentation 2008-07-18
A Study on Multiple Sound Source Localization using Extended Histogram
Takanori SOMEYA, Kenji SUYAMA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we propose a new method for multiple sound source localization utilizing weighted reliability histogram based on sparseness of speech signals. In the propose method, multiple time-frequency region are constructed by using correlation matrices corresponding to multiple forget factors. In each time-frequency points on such the regions, the direction-of-arrival (DOA) of source is estimated by using the Root-MUSIC method. Then, the extended histogram is made by weighting the reliability for each the results. Finally, the mixed Cauchy distribution is fitted to the histogram in a criterion of maximum likelihood. Several experimental results in actual room environment are shown to present an ability of multiple source localization.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) sound source localization / microphone array
Paper # EA2008-38
Date of Issue

Conference Information
Committee EA
Conference Date 2008/7/11(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 Engineering Acoustics (EA)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Study on Multiple Sound Source Localization using Extended Histogram
Sub Title (in English)
Keyword(1) sound source localization
Keyword(2) microphone array
1st Author's Name Takanori SOMEYA
1st Author's Affiliation Department of Electrical and Electronic Engineering, School of Engineering, Tokyo Denki University()
2nd Author's Name Kenji SUYAMA
2nd Author's Affiliation Department of Electrical and Electronic Engineering, School of Engineering, Tokyo Denki University
Date 2008-07-18
Paper # EA2008-38
Volume (vol) vol.108
Number (no) 143
Page pp.pp.-
#Pages 6
Date of Issue