Simultaneous processing of sound source separation and musical instrument identification using Bayesian spectral modeling

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

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

10 Citations (Scopus)

Abstract

This paper presents a method of both separating audio mixtures into sound sources and identifying the musical instruments of the sources. A statistical tone model of the power spectrogram, called an integrated model, is defined and source separation and instrument identification are carried out on the basis of Bayesian inference. Since, the parameter distributions of the integrated model depend on each instrument, the instrument name is identified by selecting the one that has the maximum relative instrument weight. Experimental results showed correct instrument identification enables precise source separation even when many overtones overlap.

Original languageEnglish
Title of host publicationICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Pages3816-3819
Number of pages4
DOIs
Publication statusPublished - 2011
Externally publishedYes
Event36th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011 - Prague
Duration: 2011 May 222011 May 27

Other

Other36th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011
CityPrague
Period11/5/2211/5/27

Fingerprint

Musical instruments
Source separation
Acoustic waves
Processing
Identification (control systems)

Keywords

  • Bayesian methods
  • instrument identification
  • Source separation
  • spectrogram

ASJC Scopus subject areas

  • Signal Processing
  • Software
  • Electrical and Electronic Engineering

Cite this

Itoyama, K., Goto, M., Komatani, K., Ogata, T., & Okuno, H. G. (2011). Simultaneous processing of sound source separation and musical instrument identification using Bayesian spectral modeling. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (pp. 3816-3819). [5947183] https://doi.org/10.1109/ICASSP.2011.5947183

Simultaneous processing of sound source separation and musical instrument identification using Bayesian spectral modeling. / Itoyama, Katsutoshi; Goto, Masataka; Komatani, Kazunori; Ogata, Tetsuya; Okuno, Hiroshi G.

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2011. p. 3816-3819 5947183.

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

Itoyama, K, Goto, M, Komatani, K, Ogata, T & Okuno, HG 2011, Simultaneous processing of sound source separation and musical instrument identification using Bayesian spectral modeling. in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings., 5947183, pp. 3816-3819, 36th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011, Prague, 11/5/22. https://doi.org/10.1109/ICASSP.2011.5947183
Itoyama K, Goto M, Komatani K, Ogata T, Okuno HG. Simultaneous processing of sound source separation and musical instrument identification using Bayesian spectral modeling. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2011. p. 3816-3819. 5947183 https://doi.org/10.1109/ICASSP.2011.5947183
Itoyama, Katsutoshi ; Goto, Masataka ; Komatani, Kazunori ; Ogata, Tetsuya ; Okuno, Hiroshi G. / Simultaneous processing of sound source separation and musical instrument identification using Bayesian spectral modeling. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2011. pp. 3816-3819
@inproceedings{4fb044801c1c4693886c1b567788066b,
title = "Simultaneous processing of sound source separation and musical instrument identification using Bayesian spectral modeling",
abstract = "This paper presents a method of both separating audio mixtures into sound sources and identifying the musical instruments of the sources. A statistical tone model of the power spectrogram, called an integrated model, is defined and source separation and instrument identification are carried out on the basis of Bayesian inference. Since, the parameter distributions of the integrated model depend on each instrument, the instrument name is identified by selecting the one that has the maximum relative instrument weight. Experimental results showed correct instrument identification enables precise source separation even when many overtones overlap.",
keywords = "Bayesian methods, instrument identification, Source separation, spectrogram",
author = "Katsutoshi Itoyama and Masataka Goto and Kazunori Komatani and Tetsuya Ogata and Okuno, {Hiroshi G.}",
year = "2011",
doi = "10.1109/ICASSP.2011.5947183",
language = "English",
isbn = "9781457705397",
pages = "3816--3819",
booktitle = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

}

TY - GEN

T1 - Simultaneous processing of sound source separation and musical instrument identification using Bayesian spectral modeling

AU - Itoyama, Katsutoshi

AU - Goto, Masataka

AU - Komatani, Kazunori

AU - Ogata, Tetsuya

AU - Okuno, Hiroshi G.

PY - 2011

Y1 - 2011

N2 - This paper presents a method of both separating audio mixtures into sound sources and identifying the musical instruments of the sources. A statistical tone model of the power spectrogram, called an integrated model, is defined and source separation and instrument identification are carried out on the basis of Bayesian inference. Since, the parameter distributions of the integrated model depend on each instrument, the instrument name is identified by selecting the one that has the maximum relative instrument weight. Experimental results showed correct instrument identification enables precise source separation even when many overtones overlap.

AB - This paper presents a method of both separating audio mixtures into sound sources and identifying the musical instruments of the sources. A statistical tone model of the power spectrogram, called an integrated model, is defined and source separation and instrument identification are carried out on the basis of Bayesian inference. Since, the parameter distributions of the integrated model depend on each instrument, the instrument name is identified by selecting the one that has the maximum relative instrument weight. Experimental results showed correct instrument identification enables precise source separation even when many overtones overlap.

KW - Bayesian methods

KW - instrument identification

KW - Source separation

KW - spectrogram

UR - http://www.scopus.com/inward/record.url?scp=80051622215&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80051622215&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2011.5947183

DO - 10.1109/ICASSP.2011.5947183

M3 - Conference contribution

SN - 9781457705397

SP - 3816

EP - 3819

BT - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

ER -