Development of Generative Music Applications having Educational Goals in Mind

We have developed a robust methodology of analysis/re-synthesis of non-Eurogenetic
musical styles for the creation of generative music applications. This methodology combines computational analysis and encoding of annotated audio data with theoretical insights about the style under consideration and its performance practice. Our approach to generative applications of music from the region represents a contribution on two levels: 1) the creation of tools for lay audiences to interact with musical heritage beyond Western music styles; and
2) the automatic generation of unlimited amounts of data for training machine learning algorithms.

A hallmark of our approach to the creation of generative music systems is the design and implementation of real-time user interaction. Examples of these applications can be found below: