Presentation 2001/2/16
Sound Source Identification Process of Polyphony using Neural Network
Jutaro Murase, Masakazu Nakanishi,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The sound source identification is to distinct what instrument the recognized sound is played by. It is one of the most important process in automatic transcription. In this paper, we have approached the sound source identification of polyphony using the neural network learned by back propagation. As for the property of sound, since it appears in harmonic structure, a harmonic relationship of the fundamental frequency component, we have used it as the feature value. As a result, we have confirmed that the harmonic structure is an important feature value, and the neural network was found to be effective for sound source identification. Moreover, since we have used only the harmonics as the feature values, a simple network was realized.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Music Information Processing / Sound Source Identification / Neural Network / Harmonic Structure
Paper # SP2000-147
Date of Issue

Conference Information
Committee SP
Conference Date 2001/2/16(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Speech (SP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Sound Source Identification Process of Polyphony using Neural Network
Sub Title (in English)
Keyword(1) Music Information Processing
Keyword(2) Sound Source Identification
Keyword(3) Neural Network
Keyword(4) Harmonic Structure
1st Author's Name Jutaro Murase
1st Author's Affiliation Graduate School of Science and Technology, Keio University()
2nd Author's Name Masakazu Nakanishi
2nd Author's Affiliation Graduate School of Science and Technology, Keio University
Date 2001/2/16
Paper # SP2000-147
Volume (vol) vol.100
Number (no) 636
Page pp.pp.-
#Pages 8
Date of Issue