Efficient query-by-content audio retrieval by locality sensitive hashing and partial sequence comparison

Yi Yu, Kazuki Joe, J. S. Downie

Research output: Contribution to journalArticle

Abstract

This paper investigates suitable indexing techniques to enable efficient content-based audio retrieval in large acoustic databases. To make an index-based retrieval mechanism applicable to audio content, we investigate the design of Locality Sensitive Hashing (LSH) and the partial sequence comparison. We propose a fast and efficient audio retrieval framework of query-by-content and develop an audio retrieval system. Based on this framework, four different audio retrieval schemes, LSH-Dynamic Programming (DP), LSH-Sparse DP (SDP), Exact Euclidian LSH (E2LSH)- DP, E2LSH-SDP, are introduced and evaluated in order to better understand the performance of audio retrieval algorithms. The experimental results indicate that compared with the traditional DP and the other three compititive schemes, E2LSH-SDP exhibits the best tradeoff in terms of the response time, retrieval accuracy and computation cost.

Original languageEnglish (US)
Pages (from-to)1730-1739
Number of pages10
JournalIEICE Transactions on Information and Systems
VolumeE91-D
Issue number6
DOIs
StatePublished - Jun 2008

Fingerprint

Dynamic programming
Acoustics
Costs

Keywords

  • Content-based audio retrieval
  • Dynamic programming
  • Indexing
  • Locality-sensitive hashing

ASJC Scopus subject areas

  • Software
  • Hardware and Architecture
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering
  • Artificial Intelligence

Cite this

Efficient query-by-content audio retrieval by locality sensitive hashing and partial sequence comparison. / Yu, Yi; Joe, Kazuki; Downie, J. S.

In: IEICE Transactions on Information and Systems, Vol. E91-D, No. 6, 06.2008, p. 1730-1739.

Research output: Contribution to journalArticle

@article{7e95df16ea4f4b999650f0cd15b26d53,
title = "Efficient query-by-content audio retrieval by locality sensitive hashing and partial sequence comparison",
abstract = "This paper investigates suitable indexing techniques to enable efficient content-based audio retrieval in large acoustic databases. To make an index-based retrieval mechanism applicable to audio content, we investigate the design of Locality Sensitive Hashing (LSH) and the partial sequence comparison. We propose a fast and efficient audio retrieval framework of query-by-content and develop an audio retrieval system. Based on this framework, four different audio retrieval schemes, LSH-Dynamic Programming (DP), LSH-Sparse DP (SDP), Exact Euclidian LSH (E2LSH)- DP, E2LSH-SDP, are introduced and evaluated in order to better understand the performance of audio retrieval algorithms. The experimental results indicate that compared with the traditional DP and the other three compititive schemes, E2LSH-SDP exhibits the best tradeoff in terms of the response time, retrieval accuracy and computation cost.",
keywords = "Content-based audio retrieval, Dynamic programming, Indexing, Locality-sensitive hashing",
author = "Yi Yu and Kazuki Joe and Downie, {J. S.}",
year = "2008",
month = "6",
doi = "10.1093/ietisy/e91-d.6.1730",
language = "English (US)",
volume = "E91-D",
pages = "1730--1739",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "6",

}

TY - JOUR

T1 - Efficient query-by-content audio retrieval by locality sensitive hashing and partial sequence comparison

AU - Yu, Yi

AU - Joe, Kazuki

AU - Downie, J. S.

PY - 2008/6

Y1 - 2008/6

N2 - This paper investigates suitable indexing techniques to enable efficient content-based audio retrieval in large acoustic databases. To make an index-based retrieval mechanism applicable to audio content, we investigate the design of Locality Sensitive Hashing (LSH) and the partial sequence comparison. We propose a fast and efficient audio retrieval framework of query-by-content and develop an audio retrieval system. Based on this framework, four different audio retrieval schemes, LSH-Dynamic Programming (DP), LSH-Sparse DP (SDP), Exact Euclidian LSH (E2LSH)- DP, E2LSH-SDP, are introduced and evaluated in order to better understand the performance of audio retrieval algorithms. The experimental results indicate that compared with the traditional DP and the other three compititive schemes, E2LSH-SDP exhibits the best tradeoff in terms of the response time, retrieval accuracy and computation cost.

AB - This paper investigates suitable indexing techniques to enable efficient content-based audio retrieval in large acoustic databases. To make an index-based retrieval mechanism applicable to audio content, we investigate the design of Locality Sensitive Hashing (LSH) and the partial sequence comparison. We propose a fast and efficient audio retrieval framework of query-by-content and develop an audio retrieval system. Based on this framework, four different audio retrieval schemes, LSH-Dynamic Programming (DP), LSH-Sparse DP (SDP), Exact Euclidian LSH (E2LSH)- DP, E2LSH-SDP, are introduced and evaluated in order to better understand the performance of audio retrieval algorithms. The experimental results indicate that compared with the traditional DP and the other three compititive schemes, E2LSH-SDP exhibits the best tradeoff in terms of the response time, retrieval accuracy and computation cost.

KW - Content-based audio retrieval

KW - Dynamic programming

KW - Indexing

KW - Locality-sensitive hashing

UR - http://www.scopus.com/inward/record.url?scp=68149163142&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=68149163142&partnerID=8YFLogxK

U2 - 10.1093/ietisy/e91-d.6.1730

DO - 10.1093/ietisy/e91-d.6.1730

M3 - Article

AN - SCOPUS:68149163142

VL - E91-D

SP - 1730

EP - 1739

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 6

ER -