Instrument equalizer for query-by-example retrieval: Improving sound source separation based on integrated harmonic and inharmonic models

Katsutoshi Itoyama, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

Research output: Chapter in Book/Report/Conference proceedingConference contribution

32 Citations (Scopus)

Abstract

This paper describes a music remixing interface, called Instrument Equalizer, that allows users to control the volume of each instrument part within existing audio recordings in real time. Although query-by-example retrieval systems need a user to prepare favorite examples (songs) in general, our interface gives a user to generate examples from existing ones by cutting or boosting some instrument/vocal parts, resulting in a variety of retrieved results. To change the volume, all instrument parts are separated from the input sound mixture using the corresponding standard MIDI file. For the separation, we used an integrated tone (timbre) model consisting of harmonic and inharmonic models that are initialized with template sounds recorded from a MIDI sound generator. The remaining but critical problem here is to deal with various performance styles and instrument bodies that are not given in the template sounds. To solve this problem, we train probabilistic distributions of timbre features by using various sounds. By adding a new constraint of maximizing the likelihood of timbre features extracted from each tone model, we succeeded in estimating model parameters that better express actual timbre.

Original languageEnglish
Title of host publicationISMIR 2008 - 9th International Conference on Music Information Retrieval
Pages133-138
Number of pages6
Publication statusPublished - 2008
Externally publishedYes
Event9th International Conference on Music Information Retrieval, ISMIR 2008 - Philadelphia, PA
Duration: 2008 Sep 142008 Sep 18

Other

Other9th International Conference on Music Information Retrieval, ISMIR 2008
CityPhiladelphia, PA
Period08/9/1408/9/18

Fingerprint

Source separation
Equalizers
Acoustic waves
Acoustic generators
Audio recordings
Sound
Harmonics
Timbre

ASJC Scopus subject areas

  • Music
  • Information Systems

Cite this

Itoyama, K., Goto, M., Komatani, K., Ogata, T., & Okuno, H. G. (2008). Instrument equalizer for query-by-example retrieval: Improving sound source separation based on integrated harmonic and inharmonic models. In ISMIR 2008 - 9th International Conference on Music Information Retrieval (pp. 133-138)

Instrument equalizer for query-by-example retrieval : Improving sound source separation based on integrated harmonic and inharmonic models. / Itoyama, Katsutoshi; Goto, Masataka; Komatani, Kazunori; Ogata, Tetsuya; Okuno, Hiroshi G.

ISMIR 2008 - 9th International Conference on Music Information Retrieval. 2008. p. 133-138.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Itoyama, K, Goto, M, Komatani, K, Ogata, T & Okuno, HG 2008, Instrument equalizer for query-by-example retrieval: Improving sound source separation based on integrated harmonic and inharmonic models. in ISMIR 2008 - 9th International Conference on Music Information Retrieval. pp. 133-138, 9th International Conference on Music Information Retrieval, ISMIR 2008, Philadelphia, PA, 08/9/14.
Itoyama K, Goto M, Komatani K, Ogata T, Okuno HG. Instrument equalizer for query-by-example retrieval: Improving sound source separation based on integrated harmonic and inharmonic models. In ISMIR 2008 - 9th International Conference on Music Information Retrieval. 2008. p. 133-138
Itoyama, Katsutoshi ; Goto, Masataka ; Komatani, Kazunori ; Ogata, Tetsuya ; Okuno, Hiroshi G. / Instrument equalizer for query-by-example retrieval : Improving sound source separation based on integrated harmonic and inharmonic models. ISMIR 2008 - 9th International Conference on Music Information Retrieval. 2008. pp. 133-138
@inproceedings{d80cadac1768463eb38f275e28e94ecf,
title = "Instrument equalizer for query-by-example retrieval: Improving sound source separation based on integrated harmonic and inharmonic models",
abstract = "This paper describes a music remixing interface, called Instrument Equalizer, that allows users to control the volume of each instrument part within existing audio recordings in real time. Although query-by-example retrieval systems need a user to prepare favorite examples (songs) in general, our interface gives a user to generate examples from existing ones by cutting or boosting some instrument/vocal parts, resulting in a variety of retrieved results. To change the volume, all instrument parts are separated from the input sound mixture using the corresponding standard MIDI file. For the separation, we used an integrated tone (timbre) model consisting of harmonic and inharmonic models that are initialized with template sounds recorded from a MIDI sound generator. The remaining but critical problem here is to deal with various performance styles and instrument bodies that are not given in the template sounds. To solve this problem, we train probabilistic distributions of timbre features by using various sounds. By adding a new constraint of maximizing the likelihood of timbre features extracted from each tone model, we succeeded in estimating model parameters that better express actual timbre.",
author = "Katsutoshi Itoyama and Masataka Goto and Kazunori Komatani and Tetsuya Ogata and Okuno, {Hiroshi G.}",
year = "2008",
language = "English",
isbn = "9780615248493",
pages = "133--138",
booktitle = "ISMIR 2008 - 9th International Conference on Music Information Retrieval",

}

TY - GEN

T1 - Instrument equalizer for query-by-example retrieval

T2 - Improving sound source separation based on integrated harmonic and inharmonic models

AU - Itoyama, Katsutoshi

AU - Goto, Masataka

AU - Komatani, Kazunori

AU - Ogata, Tetsuya

AU - Okuno, Hiroshi G.

PY - 2008

Y1 - 2008

N2 - This paper describes a music remixing interface, called Instrument Equalizer, that allows users to control the volume of each instrument part within existing audio recordings in real time. Although query-by-example retrieval systems need a user to prepare favorite examples (songs) in general, our interface gives a user to generate examples from existing ones by cutting or boosting some instrument/vocal parts, resulting in a variety of retrieved results. To change the volume, all instrument parts are separated from the input sound mixture using the corresponding standard MIDI file. For the separation, we used an integrated tone (timbre) model consisting of harmonic and inharmonic models that are initialized with template sounds recorded from a MIDI sound generator. The remaining but critical problem here is to deal with various performance styles and instrument bodies that are not given in the template sounds. To solve this problem, we train probabilistic distributions of timbre features by using various sounds. By adding a new constraint of maximizing the likelihood of timbre features extracted from each tone model, we succeeded in estimating model parameters that better express actual timbre.

AB - This paper describes a music remixing interface, called Instrument Equalizer, that allows users to control the volume of each instrument part within existing audio recordings in real time. Although query-by-example retrieval systems need a user to prepare favorite examples (songs) in general, our interface gives a user to generate examples from existing ones by cutting or boosting some instrument/vocal parts, resulting in a variety of retrieved results. To change the volume, all instrument parts are separated from the input sound mixture using the corresponding standard MIDI file. For the separation, we used an integrated tone (timbre) model consisting of harmonic and inharmonic models that are initialized with template sounds recorded from a MIDI sound generator. The remaining but critical problem here is to deal with various performance styles and instrument bodies that are not given in the template sounds. To solve this problem, we train probabilistic distributions of timbre features by using various sounds. By adding a new constraint of maximizing the likelihood of timbre features extracted from each tone model, we succeeded in estimating model parameters that better express actual timbre.

UR - http://www.scopus.com/inward/record.url?scp=80052051446&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80052051446&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:80052051446

SN - 9780615248493

SP - 133

EP - 138

BT - ISMIR 2008 - 9th International Conference on Music Information Retrieval

ER -