LNAI 2903 Applications of Soft Computing for
Musical Instrument Classification 1st Edition by
Daniel Piccoli, Mark Abernethy, Shri Rai, Shamim
Khan ISBN 9783540200574 354020057X pdf download
https://ebookball.com/product/lnai-2903-applications-of-soft-
computing-for-musical-instrument-classification-1st-edition-by-
daniel-piccoli-mark-abernethy-shri-rai-shamim-khan-
isbn-9783540200574-354020057x-11440/
Explore and download more ebooks or textbooks
at ebookball.com
, Get Your Digital Files Instantly: PDF, ePub, MOBI and More
Quick Digital Downloads: PDF, ePub, MOBI and Other Formats
LNAI 2903 A Defeasible Logic of Policy Based Intention 1st Edition by
Guido Governatori, Vineet Padmanabhan ISBN 9783540200574 354020057X
https://ebookball.com/product/lnai-2903-a-defeasible-logic-of-
policy-based-intention-1st-edition-by-guido-governatori-vineet-
padmanabhan-isbn-9783540200574-354020057x-9684/
LNAI 2903 A Tableaux System for Deontic Interpreted Systems 1st
Edition by Guido Governatori, Alessio Lomuscio, Marek Sergot ISBN
9783540200574 354020057X
https://ebookball.com/product/lnai-2903-a-tableaux-system-for-
deontic-interpreted-systems-1st-edition-by-guido-governatori-
alessio-lomuscio-marek-sergot-isbn-9783540200574-354020057x-9682/
LNAI 2903 Constructive Plausible Logic Is Relatively Consistent 1st
Edition by David Billington, Andrew Rock ISBN 9783540200574 354020057X
https://ebookball.com/product/lnai-2903-constructive-plausible-
logic-is-relatively-consistent-1st-edition-by-david-billington-
andrew-rock-isbn-9783540200574-354020057x-11392/
LNAI 2903 Design and Implementation of an Intelligent Information
Infrastructure 1st Edition by Henry Lau, Andrew Ning, Peggy Fung ISBN
9783540200574 354020057X
https://ebookball.com/product/lnai-2903-design-and-
implementation-of-an-intelligent-information-infrastructure-1st-
edition-by-henry-lau-andrew-ning-peggy-fung-
isbn-9783540200574-354020057x-9392/
,LNAI 2903 Applications of the Ecological Visualization System Using
Artificial Neural Network and Mathematical Analysis 1st Edition by Bok
Suk Shin, Cheol Ki Kim, Eui Young Cha ISBN 9783540200574 354020057X
https://ebookball.com/product/lnai-2903-applications-of-the-
ecological-visualization-system-using-artificial-neural-network-
and-mathematical-analysis-1st-edition-by-bok-suk-shin-cheol-ki-
kim-eui-young-cha-isbn-9783540200574-354020/
LNAI 2801 Developmental Neural Networks for Agents 1st Edition by Andy
Balaam ISBN 9783540200574 354020057X
https://ebookball.com/product/lnai-2801-developmental-neural-
networks-for-agents-1st-edition-by-andy-balaam-
isbn-9783540200574-354020057x-13726/
LNAI 2903 A Proposal of an Efficient Crossover Using Fitness
Prediction and Its Application 1st Edition by Atsuko Mutoh, Tsuyoshi
Nakamura, Shohei Kato, Hidenori Itoh ISBN 9783540200574 354020057X
https://ebookball.com/product/lnai-2903-a-proposal-of-an-
efficient-crossover-using-fitness-prediction-and-its-
application-1st-edition-by-atsuko-mutoh-tsuyoshi-nakamura-shohei-
kato-hidenori-itoh-isbn-9783540200574-354020057x-13466/
LNAI 2801 An Imitation Game for Emerging Action Categories 1st Edition
by Bart Jansen ISBN 9783540200574 354020057X
https://ebookball.com/product/lnai-2801-an-imitation-game-for-
emerging-action-categories-1st-edition-by-bart-jansen-
isbn-9783540200574-354020057x-13286/
LNAI 2903 MML Classification of Music Genres 1st Edition by Adrian
Bickerstaffe, Enes Makalic ISBN 9783540206460 354020646X
https://ebookball.com/product/lnai-2903-mml-classification-of-
music-genres-1st-edition-by-adrian-bickerstaffe-enes-makalic-
isbn-9783540206460-354020646x-14280/
, Applications of Soft Computing
for Musical Instrument Classification
Daniel Piccoli, Mark Abernethy, Shri Rai, and Shamim Khan
Murdoch University, Western Australia
{dpiccoli,mark.abernethy,smr,s.khan}@murdoch.edu.au
Abstract. In this paper, a method for pitch independent musical in-
strument recognition using artificial neural networks is presented. Spec-
tral features including FFT coefficients, harmonic envelopes and cepstral
coefficients are used to represent the musical instrument sounds for clas-
sification. The effectiveness of these features are compared by testing
the performance of ANNs trained with each feature. Multi-layer percep-
trons are also compared with Time-delay neural networks. The testing
and training sets both consist of fifteen note samples per musical instru-
ment within the chromatic scale from C3 to C6. Both sets consist of nine
instruments from the string, brass and woodwind families. Best results
were achieved with cepstrum coefficients with a classification accuracy
of 88 percent using a time-delay neural network, which is on par with
recent results using several different features.
Keywords: neural networks, musical instrument recognition
1 Introduction
With the advent of digital multimedia, there is an increasing need to be able
to catalogue audio data in much the same way that books are catalogued. Most
digital audio formats in use today such as MP3 and WAV contain limited meta-
data about the actual recordings that they contain [1]. However, the MPEG-7
specification requires that meta-data, such as the types of musical instruments
in a recording, should be stored in the file with the recording to enable effective
cataloguing of files [2]. The classification and identification of important features
of musical instruments in digital audio will be a step towards such a cataloguing
system. The process of classification based on a set of features is often referred
to as ’Content-Based Classification [1].’
Meta-data generated by such a system may be used by a search engine to
allow users to find specific styles of music. For example, a music teacher may be
interested in searching for audio files with certain instruments playing or music
from a certain genre. Currently, musical search systems only have the ability to
classify their music based on filenames.
The model discussed in this paper concentrates on identifying musical instru-
ments in sound recordings whilst assessing the usefulness of different features
T.D. Gedeon and L.C.C. Fung (Eds.): AI 2003, LNAI 2903, pp. 878–889, 2003.
c Springer-Verlag Berlin Heidelberg 2003
Musical Instrument Classification 1st Edition by
Daniel Piccoli, Mark Abernethy, Shri Rai, Shamim
Khan ISBN 9783540200574 354020057X pdf download
https://ebookball.com/product/lnai-2903-applications-of-soft-
computing-for-musical-instrument-classification-1st-edition-by-
daniel-piccoli-mark-abernethy-shri-rai-shamim-khan-
isbn-9783540200574-354020057x-11440/
Explore and download more ebooks or textbooks
at ebookball.com
, Get Your Digital Files Instantly: PDF, ePub, MOBI and More
Quick Digital Downloads: PDF, ePub, MOBI and Other Formats
LNAI 2903 A Defeasible Logic of Policy Based Intention 1st Edition by
Guido Governatori, Vineet Padmanabhan ISBN 9783540200574 354020057X
https://ebookball.com/product/lnai-2903-a-defeasible-logic-of-
policy-based-intention-1st-edition-by-guido-governatori-vineet-
padmanabhan-isbn-9783540200574-354020057x-9684/
LNAI 2903 A Tableaux System for Deontic Interpreted Systems 1st
Edition by Guido Governatori, Alessio Lomuscio, Marek Sergot ISBN
9783540200574 354020057X
https://ebookball.com/product/lnai-2903-a-tableaux-system-for-
deontic-interpreted-systems-1st-edition-by-guido-governatori-
alessio-lomuscio-marek-sergot-isbn-9783540200574-354020057x-9682/
LNAI 2903 Constructive Plausible Logic Is Relatively Consistent 1st
Edition by David Billington, Andrew Rock ISBN 9783540200574 354020057X
https://ebookball.com/product/lnai-2903-constructive-plausible-
logic-is-relatively-consistent-1st-edition-by-david-billington-
andrew-rock-isbn-9783540200574-354020057x-11392/
LNAI 2903 Design and Implementation of an Intelligent Information
Infrastructure 1st Edition by Henry Lau, Andrew Ning, Peggy Fung ISBN
9783540200574 354020057X
https://ebookball.com/product/lnai-2903-design-and-
implementation-of-an-intelligent-information-infrastructure-1st-
edition-by-henry-lau-andrew-ning-peggy-fung-
isbn-9783540200574-354020057x-9392/
,LNAI 2903 Applications of the Ecological Visualization System Using
Artificial Neural Network and Mathematical Analysis 1st Edition by Bok
Suk Shin, Cheol Ki Kim, Eui Young Cha ISBN 9783540200574 354020057X
https://ebookball.com/product/lnai-2903-applications-of-the-
ecological-visualization-system-using-artificial-neural-network-
and-mathematical-analysis-1st-edition-by-bok-suk-shin-cheol-ki-
kim-eui-young-cha-isbn-9783540200574-354020/
LNAI 2801 Developmental Neural Networks for Agents 1st Edition by Andy
Balaam ISBN 9783540200574 354020057X
https://ebookball.com/product/lnai-2801-developmental-neural-
networks-for-agents-1st-edition-by-andy-balaam-
isbn-9783540200574-354020057x-13726/
LNAI 2903 A Proposal of an Efficient Crossover Using Fitness
Prediction and Its Application 1st Edition by Atsuko Mutoh, Tsuyoshi
Nakamura, Shohei Kato, Hidenori Itoh ISBN 9783540200574 354020057X
https://ebookball.com/product/lnai-2903-a-proposal-of-an-
efficient-crossover-using-fitness-prediction-and-its-
application-1st-edition-by-atsuko-mutoh-tsuyoshi-nakamura-shohei-
kato-hidenori-itoh-isbn-9783540200574-354020057x-13466/
LNAI 2801 An Imitation Game for Emerging Action Categories 1st Edition
by Bart Jansen ISBN 9783540200574 354020057X
https://ebookball.com/product/lnai-2801-an-imitation-game-for-
emerging-action-categories-1st-edition-by-bart-jansen-
isbn-9783540200574-354020057x-13286/
LNAI 2903 MML Classification of Music Genres 1st Edition by Adrian
Bickerstaffe, Enes Makalic ISBN 9783540206460 354020646X
https://ebookball.com/product/lnai-2903-mml-classification-of-
music-genres-1st-edition-by-adrian-bickerstaffe-enes-makalic-
isbn-9783540206460-354020646x-14280/
, Applications of Soft Computing
for Musical Instrument Classification
Daniel Piccoli, Mark Abernethy, Shri Rai, and Shamim Khan
Murdoch University, Western Australia
{dpiccoli,mark.abernethy,smr,s.khan}@murdoch.edu.au
Abstract. In this paper, a method for pitch independent musical in-
strument recognition using artificial neural networks is presented. Spec-
tral features including FFT coefficients, harmonic envelopes and cepstral
coefficients are used to represent the musical instrument sounds for clas-
sification. The effectiveness of these features are compared by testing
the performance of ANNs trained with each feature. Multi-layer percep-
trons are also compared with Time-delay neural networks. The testing
and training sets both consist of fifteen note samples per musical instru-
ment within the chromatic scale from C3 to C6. Both sets consist of nine
instruments from the string, brass and woodwind families. Best results
were achieved with cepstrum coefficients with a classification accuracy
of 88 percent using a time-delay neural network, which is on par with
recent results using several different features.
Keywords: neural networks, musical instrument recognition
1 Introduction
With the advent of digital multimedia, there is an increasing need to be able
to catalogue audio data in much the same way that books are catalogued. Most
digital audio formats in use today such as MP3 and WAV contain limited meta-
data about the actual recordings that they contain [1]. However, the MPEG-7
specification requires that meta-data, such as the types of musical instruments
in a recording, should be stored in the file with the recording to enable effective
cataloguing of files [2]. The classification and identification of important features
of musical instruments in digital audio will be a step towards such a cataloguing
system. The process of classification based on a set of features is often referred
to as ’Content-Based Classification [1].’
Meta-data generated by such a system may be used by a search engine to
allow users to find specific styles of music. For example, a music teacher may be
interested in searching for audio files with certain instruments playing or music
from a certain genre. Currently, musical search systems only have the ability to
classify their music based on filenames.
The model discussed in this paper concentrates on identifying musical instru-
ments in sound recordings whilst assessing the usefulness of different features
T.D. Gedeon and L.C.C. Fung (Eds.): AI 2003, LNAI 2903, pp. 878–889, 2003.
c Springer-Verlag Berlin Heidelberg 2003
