Nearest Centroid Error Clustering for radial/elliptical basis function neural networks in timbre classification

Tae Hong Park, Perry Cook

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

Abstract

This paper presents a neural network approach for classification of musical instrument sounds through Radial and Elliptical Basic Functions. In particular, we discuss a novel automatic network fine-tuning method called Nearest Centroid Error Clustering (NCC) which determines a robust number of centroids for improved system performance. 829 monophonic sound examples from the string, brass, and woodwind families were used. A number of different performance techniques, dynamics, and pitches were utilized in training and testing the system resulting in 71% correct individual instrument classification (12 classes) and 88% correct instrument family (3 classes) classification.

Original languageEnglish (US)
Title of host publicationInternational Computer Music Conference, ICMC 2005
PublisherInternational Computer Music Association
StatePublished - 2005
EventInternational Computer Music Conference, ICMC 2005 - Barcelona, Spain
Duration: Sep 5 2005Sep 9 2005

Other

OtherInternational Computer Music Conference, ICMC 2005
CountrySpain
CityBarcelona
Period9/5/059/9/05

Fingerprint

Neural networks
Acoustic waves
Musical instruments
Brass
Tuning
Testing
Timbre
Neural Networks
Sound
Strings
Monophonic
Musical Instruments
Woodwind

ASJC Scopus subject areas

  • Computer Science Applications
  • Media Technology
  • Music

Cite this

Park, T. H., & Cook, P. (2005). Nearest Centroid Error Clustering for radial/elliptical basis function neural networks in timbre classification. In International Computer Music Conference, ICMC 2005 International Computer Music Association.

Nearest Centroid Error Clustering for radial/elliptical basis function neural networks in timbre classification. / Park, Tae Hong; Cook, Perry.

International Computer Music Conference, ICMC 2005. International Computer Music Association, 2005.

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

Park, TH & Cook, P 2005, Nearest Centroid Error Clustering for radial/elliptical basis function neural networks in timbre classification. in International Computer Music Conference, ICMC 2005. International Computer Music Association, International Computer Music Conference, ICMC 2005, Barcelona, Spain, 9/5/05.
Park TH, Cook P. Nearest Centroid Error Clustering for radial/elliptical basis function neural networks in timbre classification. In International Computer Music Conference, ICMC 2005. International Computer Music Association. 2005
Park, Tae Hong ; Cook, Perry. / Nearest Centroid Error Clustering for radial/elliptical basis function neural networks in timbre classification. International Computer Music Conference, ICMC 2005. International Computer Music Association, 2005.
@inproceedings{43f01bf120874dfbbeb5e57c6857ad40,
title = "Nearest Centroid Error Clustering for radial/elliptical basis function neural networks in timbre classification",
abstract = "This paper presents a neural network approach for classification of musical instrument sounds through Radial and Elliptical Basic Functions. In particular, we discuss a novel automatic network fine-tuning method called Nearest Centroid Error Clustering (NCC) which determines a robust number of centroids for improved system performance. 829 monophonic sound examples from the string, brass, and woodwind families were used. A number of different performance techniques, dynamics, and pitches were utilized in training and testing the system resulting in 71{\%} correct individual instrument classification (12 classes) and 88{\%} correct instrument family (3 classes) classification.",
author = "Park, {Tae Hong} and Perry Cook",
year = "2005",
language = "English (US)",
booktitle = "International Computer Music Conference, ICMC 2005",
publisher = "International Computer Music Association",

}

TY - GEN

T1 - Nearest Centroid Error Clustering for radial/elliptical basis function neural networks in timbre classification

AU - Park, Tae Hong

AU - Cook, Perry

PY - 2005

Y1 - 2005

N2 - This paper presents a neural network approach for classification of musical instrument sounds through Radial and Elliptical Basic Functions. In particular, we discuss a novel automatic network fine-tuning method called Nearest Centroid Error Clustering (NCC) which determines a robust number of centroids for improved system performance. 829 monophonic sound examples from the string, brass, and woodwind families were used. A number of different performance techniques, dynamics, and pitches were utilized in training and testing the system resulting in 71% correct individual instrument classification (12 classes) and 88% correct instrument family (3 classes) classification.

AB - This paper presents a neural network approach for classification of musical instrument sounds through Radial and Elliptical Basic Functions. In particular, we discuss a novel automatic network fine-tuning method called Nearest Centroid Error Clustering (NCC) which determines a robust number of centroids for improved system performance. 829 monophonic sound examples from the string, brass, and woodwind families were used. A number of different performance techniques, dynamics, and pitches were utilized in training and testing the system resulting in 71% correct individual instrument classification (12 classes) and 88% correct instrument family (3 classes) classification.

UR - http://www.scopus.com/inward/record.url?scp=84923863381&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84923863381&partnerID=8YFLogxK

M3 - Conference contribution

BT - International Computer Music Conference, ICMC 2005

PB - International Computer Music Association

ER -