Music is widely used for various purposes, it contributes to the treatment of patients since people appreciate it and use it for music. Music signal processing researchers are expected to propose more efficient and effective solution for real world. This paper proposed a multi-resolution feature extraction technique to deal with music signals. I utilized a fuzzy system to search their clusters of features. The fuzzy rules with variable fuzzy regions were defined by activation rectangles, which show the existence region of data for a class and overlapping rectangles, which overlapping the existence of the data for the other classes. The effectiveness of the clusters is encouraging.