We present AMIGO, a real-time computer music system that assists novice users in the composition process through guided musical improvisation. The system consists of 1) a computational analysis-generation algorithm, which not only formalizes musical principles from examples, but also guides the user in selecting note sequences; 2) a MIDI keyboard controller with an integrated LED stripe, which provides visual feedback to the user; and 3) a real-time music notation, which displays the generated output. Ultimately, AMIGO allows the intuitive creation of new musical structures and the acquisition of Western music formalisms, such as musical notation.
(an extension of AMIGO that integrates MyJazzBand)
The proposed DMI integrates two existing technologies: the interface of AMIGO, which aims to assist novice users in the composition via a light-up physical keyboard interface and the generative software component of MyJazzBand (Dias, 2018), an installation that invites users to experience jazz improvisation without prior knowledge of musical theory and practice. Inspired by an enactivist approach, we conducted an informal experiment with accomplished instrumentalists, to discuss future iterations of our DMI in supporting an intuitive understanding of jazz improvisation.
The DMI includes two major components: an illuminated keyboard controller (hardware) and a generative engine (software). Note selection is guided by visual feedback, which highlights note scales complying with a given harmonic structure. The generative nature of the engine allows high degrees of novelty while guaranteeing a structurally-coherent harmonic structure anchored in the blues/jazz idiom. Preliminary experimental results inform critical directions for future design iterations of the proposed system.
Dias. R.: Interfacing Jazz: A Study in Computer-Mediated Jazz Music Creation and Performance. Ph.D. Thesis, University of Porto, Portugal, 2018.