Using machine-learning methods for musical style modeling

Constructing a musical theory from examples presents an intellectual challenge that could foster a range of new creative applications. Thus, the authors sought to apply machine-learning methods to the problem of musical style modeling. Their work has produced examples of musical generation and applications to a computer-aided composition system. Using statistical and information-theoretic tools that analyze musical pieces, they seek to capture some of the regularity apparent in the composition process. The resulting models can be used for inference and prediction; and to generate new works that imitate the great masters' styles.