An efficient audio fingerprint search algorithm for music retrieval

Sunhyung Lee, Dongsuk Yook, Sukmoon Chang

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

The conventional audio fingerprinting system by Haitsma uses a lookup table to identify the candidate songs in the database, which contains the sub-fingerprints of songs, and searches the candidates to find a song whose bit error rate is the lowest. However, this approach has a drawback that the number of database accesses increases dramatically, especially when the database contains a large number of songs or when a matching sub-fingerprint is not found in the lookup table due to a heavily degraded input signal. In this paper, a novel search method is proposed to overcome these difficulties. The proposed method partitions each song found from the lookup table into blocks, assigns a weight to each block, and uses the weight as a search priority to speed up the search process while reducing the number of database accesses. Various results from our experiment show the significant improvement in search speed while maintaining the search accuracy comparable to the conventional method.

Original languageEnglish (US)
Article number6626252
Pages (from-to)652-656
Number of pages5
JournalIEEE Transactions on Consumer Electronics
Volume59
Issue number3
DOIs
StatePublished - Oct 31 2013

Fingerprint

Table lookup
Audio systems
Bit error rate
Experiments

All Science Journal Classification (ASJC) codes

  • Media Technology
  • Electrical and Electronic Engineering

Cite this

@article{a539497e4d2d4ea3a71e5429eb5fc4ad,
title = "An efficient audio fingerprint search algorithm for music retrieval",
abstract = "The conventional audio fingerprinting system by Haitsma uses a lookup table to identify the candidate songs in the database, which contains the sub-fingerprints of songs, and searches the candidates to find a song whose bit error rate is the lowest. However, this approach has a drawback that the number of database accesses increases dramatically, especially when the database contains a large number of songs or when a matching sub-fingerprint is not found in the lookup table due to a heavily degraded input signal. In this paper, a novel search method is proposed to overcome these difficulties. The proposed method partitions each song found from the lookup table into blocks, assigns a weight to each block, and uses the weight as a search priority to speed up the search process while reducing the number of database accesses. Various results from our experiment show the significant improvement in search speed while maintaining the search accuracy comparable to the conventional method.",
author = "Sunhyung Lee and Dongsuk Yook and Sukmoon Chang",
year = "2013",
month = "10",
day = "31",
doi = "10.1109/TCE.2013.6626252",
language = "English (US)",
volume = "59",
pages = "652--656",
journal = "IEEE Transactions on Consumer Electronics",
issn = "0098-3063",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "3",

}

An efficient audio fingerprint search algorithm for music retrieval. / Lee, Sunhyung; Yook, Dongsuk; Chang, Sukmoon.

In: IEEE Transactions on Consumer Electronics, Vol. 59, No. 3, 6626252, 31.10.2013, p. 652-656.

Research output: Contribution to journalArticle

TY - JOUR

T1 - An efficient audio fingerprint search algorithm for music retrieval

AU - Lee, Sunhyung

AU - Yook, Dongsuk

AU - Chang, Sukmoon

PY - 2013/10/31

Y1 - 2013/10/31

N2 - The conventional audio fingerprinting system by Haitsma uses a lookup table to identify the candidate songs in the database, which contains the sub-fingerprints of songs, and searches the candidates to find a song whose bit error rate is the lowest. However, this approach has a drawback that the number of database accesses increases dramatically, especially when the database contains a large number of songs or when a matching sub-fingerprint is not found in the lookup table due to a heavily degraded input signal. In this paper, a novel search method is proposed to overcome these difficulties. The proposed method partitions each song found from the lookup table into blocks, assigns a weight to each block, and uses the weight as a search priority to speed up the search process while reducing the number of database accesses. Various results from our experiment show the significant improvement in search speed while maintaining the search accuracy comparable to the conventional method.

AB - The conventional audio fingerprinting system by Haitsma uses a lookup table to identify the candidate songs in the database, which contains the sub-fingerprints of songs, and searches the candidates to find a song whose bit error rate is the lowest. However, this approach has a drawback that the number of database accesses increases dramatically, especially when the database contains a large number of songs or when a matching sub-fingerprint is not found in the lookup table due to a heavily degraded input signal. In this paper, a novel search method is proposed to overcome these difficulties. The proposed method partitions each song found from the lookup table into blocks, assigns a weight to each block, and uses the weight as a search priority to speed up the search process while reducing the number of database accesses. Various results from our experiment show the significant improvement in search speed while maintaining the search accuracy comparable to the conventional method.

UR - http://www.scopus.com/inward/record.url?scp=84886544613&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84886544613&partnerID=8YFLogxK

U2 - 10.1109/TCE.2013.6626252

DO - 10.1109/TCE.2013.6626252

M3 - Article

AN - SCOPUS:84886544613

VL - 59

SP - 652

EP - 656

JO - IEEE Transactions on Consumer Electronics

JF - IEEE Transactions on Consumer Electronics

SN - 0098-3063

IS - 3

M1 - 6626252

ER -