In this paper we investigate the use of pattern recognition techniques for recognizing two features of Persian music, dastgah and maqam, by their relation to musical scale and mode. We define some statistical measures that characterize the melodic pattern of a musical signal. We then use a standard perceptron for pattern recognition. This technique could be generalized for recognizing such features in music of other nations and also used in applications like content-based search and indexing of musical databases.
Download Full PDF Version (Non-Commercial Use)