An efficient frequent melody indexing method to improve the performance of query-by-humming systems

Jinhee You, Sang Hyun Park, Inbum Kim

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

In recent years, the need to efficiently store and retrieve large amounts of musical information has increased. In this paper, we design and implement a Query-By-Humming (QBH) system, which can retrieve melodies similar to users' humming. To make this QBH system efficient, the following three methods were proposed. First, we convert the melodies to be indexed into the corresponding strings, in order to increase search speed. The conversion method is designed to tolerate the errors involved in humming. Second, we extract significant melodies from music and then build a couple of indexes from them. For this task, we propose reliable methods for extracting melodies that occur frequently and for melodies that begin after a long rest. Third, we propose a three-step index searching method for minimizing database access. Through the experiments with a real-world data set, it was verified that this system has noticeable improvements over the N-gram approach.

Original languageEnglish
Pages (from-to)777-798
Number of pages22
JournalJournal of Information Science
Volume34
Issue number6
DOIs
Publication statusPublished - 2008 Dec 1

Fingerprint

indexing
performance
Experiments
music
experiment

All Science Journal Classification (ASJC) codes

  • Information Systems
  • Library and Information Sciences

Cite this

@article{3578ef8a342a4a19ba7f1f13161c8f0f,
title = "An efficient frequent melody indexing method to improve the performance of query-by-humming systems",
abstract = "In recent years, the need to efficiently store and retrieve large amounts of musical information has increased. In this paper, we design and implement a Query-By-Humming (QBH) system, which can retrieve melodies similar to users' humming. To make this QBH system efficient, the following three methods were proposed. First, we convert the melodies to be indexed into the corresponding strings, in order to increase search speed. The conversion method is designed to tolerate the errors involved in humming. Second, we extract significant melodies from music and then build a couple of indexes from them. For this task, we propose reliable methods for extracting melodies that occur frequently and for melodies that begin after a long rest. Third, we propose a three-step index searching method for minimizing database access. Through the experiments with a real-world data set, it was verified that this system has noticeable improvements over the N-gram approach.",
author = "Jinhee You and Park, {Sang Hyun} and Inbum Kim",
year = "2008",
month = "12",
day = "1",
doi = "10.1177/0165551507087712",
language = "English",
volume = "34",
pages = "777--798",
journal = "Journal of Information Science",
issn = "0165-5515",
publisher = "SAGE Publications Ltd",
number = "6",

}

An efficient frequent melody indexing method to improve the performance of query-by-humming systems. / You, Jinhee; Park, Sang Hyun; Kim, Inbum.

In: Journal of Information Science, Vol. 34, No. 6, 01.12.2008, p. 777-798.

Research output: Contribution to journalArticle

TY - JOUR

T1 - An efficient frequent melody indexing method to improve the performance of query-by-humming systems

AU - You, Jinhee

AU - Park, Sang Hyun

AU - Kim, Inbum

PY - 2008/12/1

Y1 - 2008/12/1

N2 - In recent years, the need to efficiently store and retrieve large amounts of musical information has increased. In this paper, we design and implement a Query-By-Humming (QBH) system, which can retrieve melodies similar to users' humming. To make this QBH system efficient, the following three methods were proposed. First, we convert the melodies to be indexed into the corresponding strings, in order to increase search speed. The conversion method is designed to tolerate the errors involved in humming. Second, we extract significant melodies from music and then build a couple of indexes from them. For this task, we propose reliable methods for extracting melodies that occur frequently and for melodies that begin after a long rest. Third, we propose a three-step index searching method for minimizing database access. Through the experiments with a real-world data set, it was verified that this system has noticeable improvements over the N-gram approach.

AB - In recent years, the need to efficiently store and retrieve large amounts of musical information has increased. In this paper, we design and implement a Query-By-Humming (QBH) system, which can retrieve melodies similar to users' humming. To make this QBH system efficient, the following three methods were proposed. First, we convert the melodies to be indexed into the corresponding strings, in order to increase search speed. The conversion method is designed to tolerate the errors involved in humming. Second, we extract significant melodies from music and then build a couple of indexes from them. For this task, we propose reliable methods for extracting melodies that occur frequently and for melodies that begin after a long rest. Third, we propose a three-step index searching method for minimizing database access. Through the experiments with a real-world data set, it was verified that this system has noticeable improvements over the N-gram approach.

UR - http://www.scopus.com/inward/record.url?scp=57249098271&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=57249098271&partnerID=8YFLogxK

U2 - 10.1177/0165551507087712

DO - 10.1177/0165551507087712

M3 - Article

VL - 34

SP - 777

EP - 798

JO - Journal of Information Science

JF - Journal of Information Science

SN - 0165-5515

IS - 6

ER -