Presentation 2015-03-05
Adaptation of Machine Learning Method for Music Structure Analysis
Yoshiyuki KUSHIBE, Toshiaki TAKITA, Masatoshi HAMANAKA, Sakurako YAZAWA, Junichi HOSHINO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper describes the music structure analysis method using machine learning. Music structure analysis is to automatically extract the musical structures, such as verse and chorus, from the audio-based music data. In the previous work, the method using self-similarity matrix is proposed. It is the matrix to convert an audio data into a suitable feature sequence, and then to compare all elements of the sequence with each other. However, the result is poor accuracy for the music with the unexpected situation. There is a lot of factor becoming the obstacle of analysis, for example, change of the tempo, noise and change of musical instrument. We propose the method to learn the music to contain their factors using Deep Belief Networks which is a kind of machine learning.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Music Structure Analysis / Deep Belief Networks / Restricted Boltzmann Machine
Paper # IBISML2014-89
Date of Issue

Conference Information
Committee IBISML
Conference Date 2015/2/26(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 Information-Based Induction Sciences and Machine Learning (IBISML)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Adaptation of Machine Learning Method for Music Structure Analysis
Sub Title (in English)
Keyword(1) Music Structure Analysis
Keyword(2) Deep Belief Networks
Keyword(3) Restricted Boltzmann Machine
1st Author's Name Yoshiyuki KUSHIBE
1st Author's Affiliation Graduate School of Systems and Information Engineering, University of Tsukuba()
2nd Author's Name Toshiaki TAKITA
2nd Author's Affiliation Graduate School of Systems and Information Engineering, University of Tsukuba
3rd Author's Name Masatoshi HAMANAKA
3rd Author's Affiliation Graduate School of Medicine, Kyoto University
4th Author's Name Sakurako YAZAWA
4th Author's Affiliation Graduate School of Systems and Information Engineering, University of Tsukuba
5th Author's Name Junichi HOSHINO
5th Author's Affiliation Graduate School of Systems and Information Engineering, University of Tsukuba
Date 2015-03-05
Paper # IBISML2014-89
Volume (vol) vol.114
Number (no) 502
Page pp.pp.-
#Pages 8
Date of Issue