The proposal in this study is to design a tone creation supporting system based on KANSEI information. The recent computer innovation brought many things that used to be difficult to realize. However, it is hard for inexperienced persons to make a tone from the present Desk Top Music Software. With in a technical term, it is hard for them to imagine what kind of tone to be heard. If technical terms are transposed to KANSEI words, we will be able to imagine a tone easier than now. This study is a challenge to build the system based on the KANSEI information from which everyone can make a tone more freely. This study uses the wavelet transform to analyze sounds, the factor analysis which used the maximum likelihood estimate method to analyze the subjectivity evaluation value acquired by the semantic differential method.