Eureka! Institutional Repository

This postgraduate thesis presents a less common approach to artifical intelligence driven music synthesis. A web application is created that allows multiple users to collaboratively generate music through an evolutionary process. The system uses an evolutionary algorithm to evolve populations of melodies that attempt to match a predefined chord progression. Users can provide feedback on the musical output, which alter system’s evaluation. More than 30 people used this platform and provided ratings. The findings of this study indicate that while it is possible to create musically pleasing melodies without extensive knowledge of complex algorithms, developing a user friendly interface and process is a more complicated task. The results suggest that the platform has the potential to facilitate collaborative music generation and could be used in various contexts, such as composition and live performances. Consequently, this study provides a basis for a potential user friendly and interactive web application for musical synthesis.