Automatic Music Composition Adapting to Personal Sensibility Based on Evolutionary Computation Algorithm

2017 Fiscal Year Final Research Report

• Project/Area Number: 26330318
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Kansei informatics
• Research Institution: Tokyo City University
• Principal Investigator: OTANI Noriko 東京都市大学, メディア情報学部, 教授 (70328566)
• Co-Investigator(Renkei-kenkyūsha): NUMAO Masayuki 大阪大学, 産業科学研究所, 教授 (30198551) ; AIBA Eriko 電気通信大学, 情報理工学系研究科, 助教 (40569761)
• Project Period (FY): 2014-04-01 – 2018-03-31
• Keywords: 自動作曲 / 感性モデル獲得 / 進化計算アルゴリズム

Outline of Final Research Achievements

In this research, we aimed to construct an automatic music composition system adapting to personal sensibility, and to provide an environment where music can be enjoyed for more various use. A method of generating chord progressions in consideration of the entire music structure, a method for extracting personal sensibility models using only existing musical pieces adapting the objective as training data, and a method of utilizing the automatic music composition system in artists' creative activities were proposed. Those effectiveness were confirmed in some experiments and case studies. In addition, we developed a browser-based interface, and held some workshops of experience of our automatic music composition system in various events.

Free Research Field

機械学習，進化計算アルゴリズム