著者
大西 厳 木村 一郎 山田 英巳
出版者
一般社団法人 日本機械学会
雑誌
日本機械学会論文集 C編 (ISSN:03875024)
巻号頁・発行日
vol.65, no.635, pp.2787-2793, 1999-07-25 (Released:2008-02-26)
参考文献数
13
被引用文献数
2 3

Recently, we have been demanding products that satisfy personal tastes. In these days, it is extremely important to make a human sensibility model and then build up a technology system for applying human sensibility to various fields in engineering. In this paper, a human sensibility model for musical chords is studied. The model that uses an artificial neural network is constructed by ccnsidering the structure of a cochlea that analyzes sounds into frequency. The musical chords that are composed of three or more musical tones are inputted into the neural network, and a set of the sensibility information that characterizes "cheerfulness-gloominess", "thickness-thinness" and "stability instability" is outputted. After training the model, the following results that agree with the knowledge of psychology and physiology are proved. 1) Some hidden units for "cheerfulness- gloominess" and "stability instability" are influenced by the critical bandwidth of musical tone frequency, and furthermore they have "Tonality". 2) For "thickness thinness", there is a unit that distinguishes musical tones by frequency.