Musical Agent Systems: MACAT and MACataRT

Our research explores the development of musical agents—artificial systems that automate creative tasks such as composition and improvisation using AI and multi-agent systems. These agents can learn, adapt, and collaborate with human musicians in real-time, employing techniques like corpus-based concatenative synthesis to maintain expressive nuances in performance. We introduce MACAT and MACataRT, two systems designed to enable interactive, human-in-the-loop generative AI to expand creative possibilities for musicians in co-creative settings.