Abstract: Music structure is not only an important form of the music works to express artists' ideas, but also is an effective way for listeners to understand the meaning of the music. This paper proposes a timbre unit modeling method based on musical features, using unsupervised clustering method to analyze music structure according to the distribution of local timbre units, with the amended Fisher rule. The experiment results show that the unsupervised clustering algorithm with DCT-based chroma feature is an effective way to analyze the music structure.