Musical instrument identification based on F0-dependent multivariate normal distribution

Tetsuro Kitahara, Masataka Goto, Hiroshi G. Okuno

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

31 Citations (Scopus)

Abstract

The pitch dependency of timbres has not been fully exploited in musical instrument identification. In this paper, we present a method using an F0-dependent multivariate normal distribution of which mean is represented by a function of fundamental frequency (F0). This F0-dependent mean function represents the pitch dependency of each feature, while the F0-normalized covariance represents the non-pitch dependency. Musical instrument sounds are first analyzed by the F0-dependent multivariate normal distribution, and then identified by using the discriminant function based on the Bayes decision rule. Experimental results of identifying 6,247 solo tones of 19 musical instruments by 10-fold cross validation showed that the proposed method improved the recognition rate at individual-instrument level from 75.73% to 79.73%, and the recognition rate at category level from 88.20% to 90.65%.

Original languageEnglish
Title of host publicationICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Pages421-424
Number of pages4
Volume5
Publication statusPublished - 2003
Externally publishedYes
Event2003 IEEE International Conference on Accoustics, Speech, and Signal Processing - Hong Kong, Hong Kong
Duration: 2003 Apr 62003 Apr 10

Other

Other2003 IEEE International Conference on Accoustics, Speech, and Signal Processing
CountryHong Kong
CityHong Kong
Period03/4/603/4/10

Fingerprint

Musical instruments
Normal distribution
normal density functions
Acoustic waves
acoustics

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Signal Processing
  • Acoustics and Ultrasonics

Cite this

Kitahara, T., Goto, M., & Okuno, H. G. (2003). Musical instrument identification based on F0-dependent multivariate normal distribution. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (Vol. 5, pp. 421-424)

Musical instrument identification based on F0-dependent multivariate normal distribution. / Kitahara, Tetsuro; Goto, Masataka; Okuno, Hiroshi G.

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. 5 2003. p. 421-424.

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

Kitahara, T, Goto, M & Okuno, HG 2003, Musical instrument identification based on F0-dependent multivariate normal distribution. in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. vol. 5, pp. 421-424, 2003 IEEE International Conference on Accoustics, Speech, and Signal Processing, Hong Kong, Hong Kong, 03/4/6.
Kitahara T, Goto M, Okuno HG. Musical instrument identification based on F0-dependent multivariate normal distribution. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. 5. 2003. p. 421-424
Kitahara, Tetsuro ; Goto, Masataka ; Okuno, Hiroshi G. / Musical instrument identification based on F0-dependent multivariate normal distribution. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. 5 2003. pp. 421-424
@inproceedings{581af3743563489097a37cc51d89d670,
title = "Musical instrument identification based on F0-dependent multivariate normal distribution",
abstract = "The pitch dependency of timbres has not been fully exploited in musical instrument identification. In this paper, we present a method using an F0-dependent multivariate normal distribution of which mean is represented by a function of fundamental frequency (F0). This F0-dependent mean function represents the pitch dependency of each feature, while the F0-normalized covariance represents the non-pitch dependency. Musical instrument sounds are first analyzed by the F0-dependent multivariate normal distribution, and then identified by using the discriminant function based on the Bayes decision rule. Experimental results of identifying 6,247 solo tones of 19 musical instruments by 10-fold cross validation showed that the proposed method improved the recognition rate at individual-instrument level from 75.73{\%} to 79.73{\%}, and the recognition rate at category level from 88.20{\%} to 90.65{\%}.",
author = "Tetsuro Kitahara and Masataka Goto and Okuno, {Hiroshi G.}",
year = "2003",
language = "English",
volume = "5",
pages = "421--424",
booktitle = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

}

TY - GEN

T1 - Musical instrument identification based on F0-dependent multivariate normal distribution

AU - Kitahara, Tetsuro

AU - Goto, Masataka

AU - Okuno, Hiroshi G.

PY - 2003

Y1 - 2003

N2 - The pitch dependency of timbres has not been fully exploited in musical instrument identification. In this paper, we present a method using an F0-dependent multivariate normal distribution of which mean is represented by a function of fundamental frequency (F0). This F0-dependent mean function represents the pitch dependency of each feature, while the F0-normalized covariance represents the non-pitch dependency. Musical instrument sounds are first analyzed by the F0-dependent multivariate normal distribution, and then identified by using the discriminant function based on the Bayes decision rule. Experimental results of identifying 6,247 solo tones of 19 musical instruments by 10-fold cross validation showed that the proposed method improved the recognition rate at individual-instrument level from 75.73% to 79.73%, and the recognition rate at category level from 88.20% to 90.65%.

AB - The pitch dependency of timbres has not been fully exploited in musical instrument identification. In this paper, we present a method using an F0-dependent multivariate normal distribution of which mean is represented by a function of fundamental frequency (F0). This F0-dependent mean function represents the pitch dependency of each feature, while the F0-normalized covariance represents the non-pitch dependency. Musical instrument sounds are first analyzed by the F0-dependent multivariate normal distribution, and then identified by using the discriminant function based on the Bayes decision rule. Experimental results of identifying 6,247 solo tones of 19 musical instruments by 10-fold cross validation showed that the proposed method improved the recognition rate at individual-instrument level from 75.73% to 79.73%, and the recognition rate at category level from 88.20% to 90.65%.

UR - http://www.scopus.com/inward/record.url?scp=0141631749&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0141631749&partnerID=8YFLogxK

M3 - Conference contribution

VL - 5

SP - 421

EP - 424

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

ER -