AbstractsComputer Science

In this thesis, two music accompaniment systems are presented. Evac (the evolutionary accompanist) is a system that engages in musical improvisation with the user. It uses a novel, implicitly interactive, genetic algorithm (GA), which allows the user’s actions to influence Evac’s musical performance without the need for explicit rating of individuals. Evac runs in real time, allowing the user to experience the same kind of exploration that happens in real life improvisation scenarios with other musicians. EvolMusic is an accompaniment system involving human preference learning. It allows direct control from the user over the accompaniment by using machine learning techniques to learn a fitness function from the user’s preferences. EvolMusic records a piece of the user’s musical input, generates different accompaniments, lets the user vote for his or her favorite, adjusts the GA’s fitness function, and then generates new accompaniments which can be further used to learn the user’s preferences.