Fast Sound Source Localization Using Two-Level Search Space Clustering

Dongsuk Yook, Taewoo Lee, Youngkyu Cho

Research output: Contribution to journalArticle

22 Citations (Scopus)

Abstract

Steered response power phase transform (SRP-PHAT) is a method that is widely used for robust sound source localization (SSL). However, since SRP-PHAT searches over a large number of candidate locations, it is too slow to run in real-time for large-scale microphone array systems. In this paper, we propose a robust two-level search space clustering method to speed-up SRP-PHAT-based SSL. The proposed method divides the candidate locations of the sound source into a set of groups and finds a small number of groups that are likely to contain the maximum power location. By searching within the small number of groups, the computational costs are reduced by 61.8% compared to a previously proposed method without loss of accuracy.

Original languageEnglish
Article number7039285
Pages (from-to)20-26
Number of pages7
JournalIEEE Transactions on Cybernetics
Volume46
Issue number1
DOIs
Publication statusPublished - 2016 Jan 1

Fingerprint

Acoustic waves
Microphones
Costs

Keywords

  • Search space clustering (SSC)
  • Sound source localization (SSL)
  • Steered response power phase transform (SRP-PHAT)

ASJC Scopus subject areas

  • Computer Science Applications
  • Human-Computer Interaction
  • Information Systems
  • Software
  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Cite this

Fast Sound Source Localization Using Two-Level Search Space Clustering. / Yook, Dongsuk; Lee, Taewoo; Cho, Youngkyu.

In: IEEE Transactions on Cybernetics, Vol. 46, No. 1, 7039285, 01.01.2016, p. 20-26.

Research output: Contribution to journalArticle

@article{0fcbecb3934940eb9663a6a958ed2a27,
title = "Fast Sound Source Localization Using Two-Level Search Space Clustering",
abstract = "Steered response power phase transform (SRP-PHAT) is a method that is widely used for robust sound source localization (SSL). However, since SRP-PHAT searches over a large number of candidate locations, it is too slow to run in real-time for large-scale microphone array systems. In this paper, we propose a robust two-level search space clustering method to speed-up SRP-PHAT-based SSL. The proposed method divides the candidate locations of the sound source into a set of groups and finds a small number of groups that are likely to contain the maximum power location. By searching within the small number of groups, the computational costs are reduced by 61.8{\%} compared to a previously proposed method without loss of accuracy.",
keywords = "Search space clustering (SSC), Sound source localization (SSL), Steered response power phase transform (SRP-PHAT)",
author = "Dongsuk Yook and Taewoo Lee and Youngkyu Cho",
year = "2016",
month = "1",
day = "1",
doi = "10.1109/TCYB.2015.2391252",
language = "English",
volume = "46",
pages = "20--26",
journal = "IEEE Transactions on Cybernetics",
issn = "2168-2267",
publisher = "IEEE Advancing Technology for Humanity",
number = "1",

}

TY - JOUR

T1 - Fast Sound Source Localization Using Two-Level Search Space Clustering

AU - Yook, Dongsuk

AU - Lee, Taewoo

AU - Cho, Youngkyu

PY - 2016/1/1

Y1 - 2016/1/1

N2 - Steered response power phase transform (SRP-PHAT) is a method that is widely used for robust sound source localization (SSL). However, since SRP-PHAT searches over a large number of candidate locations, it is too slow to run in real-time for large-scale microphone array systems. In this paper, we propose a robust two-level search space clustering method to speed-up SRP-PHAT-based SSL. The proposed method divides the candidate locations of the sound source into a set of groups and finds a small number of groups that are likely to contain the maximum power location. By searching within the small number of groups, the computational costs are reduced by 61.8% compared to a previously proposed method without loss of accuracy.

AB - Steered response power phase transform (SRP-PHAT) is a method that is widely used for robust sound source localization (SSL). However, since SRP-PHAT searches over a large number of candidate locations, it is too slow to run in real-time for large-scale microphone array systems. In this paper, we propose a robust two-level search space clustering method to speed-up SRP-PHAT-based SSL. The proposed method divides the candidate locations of the sound source into a set of groups and finds a small number of groups that are likely to contain the maximum power location. By searching within the small number of groups, the computational costs are reduced by 61.8% compared to a previously proposed method without loss of accuracy.

KW - Search space clustering (SSC)

KW - Sound source localization (SSL)

KW - Steered response power phase transform (SRP-PHAT)

UR - http://www.scopus.com/inward/record.url?scp=84960340932&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84960340932&partnerID=8YFLogxK

U2 - 10.1109/TCYB.2015.2391252

DO - 10.1109/TCYB.2015.2391252

M3 - Article

AN - SCOPUS:84960340932

VL - 46

SP - 20

EP - 26

JO - IEEE Transactions on Cybernetics

JF - IEEE Transactions on Cybernetics

SN - 2168-2267

IS - 1

M1 - 7039285

ER -