Robert Keller: ‘Jamming with Improvisor’

Impro-Visor (Improvisation Advisor) is a free, open-source, program designed by the performer and his students to be an improvisational companion for private study. Although not primarily intended for public performance, here we explore that possibility by demonstrating a user trading with Impro-Visor. Trading is a concept that occurs in typical jazz performances wherein musicians alternate playing a number of measures, responding to each other. Most common is one or more melody instruments trading four or eight bars with a percussion instrument, although here we will have one melody instrument played by the performer trading with the Impro-Visor program as the responder. Numerous types of response are possible, but I will concentrate on the use of transformational grammars.

Robert M. Keller has been a professor of computer science at Harvey Mudd College, Claremont, California, USA since 1991. He is also a jazz musician and teaches jazz improvisation to students of the five colleges in the Claremont Consortium. Previous faculty appointments include Princeton University, the University of Utah, and University of California, Davis. His degrees are from Washington University in Saint Louis and the University of California, Berkeley.
Since 2005, Professor Keller has lead the Intelligent Music Software project, which produces, with the help of approximately sixty undergraduate students, the Impro-Visor (Improvisation Advisor) free and open-source software, which is designed to help musicians learn to improvise or improve their improvisational skills. In addition to computer music and computational creativity, his publications include work in program verification, functional programming, parallel computing and architecture, information retrieval, and neural networks.

René Mogensen – ‘Favoleggiatori 2’

‘Favoleggiatori 2’ is a computer system that includes ‘computational creativity‘ as the basis for a machine improviser. The machine co-improvises together with a human saxophonist and as such the performance may be considered a duo of human and artificial musical intelligence. The internal activities of this machine improviser include artificial swarms operating in dynamic topological spaces. The topological spaces represent memory of the performed music and are changed by saxophonist’s musical performance during the piece. The swarm agent activities within the topological spaces are mapped onto additive synthesis, MIDI instruments, pre-recorded audio samples, and spatialisation parameters which constitute the output of the machine improviser.

René Mogensen (PhD, MA, MM, BA) teaches creative use of technology in music composition and performance, as well as contemporary musical concepts and techniques, at the Royal Birmingham Conservatoire, UK. Much of his current research includes work on computational creativity in music as well as work on analysis of mixed music which integrates acoustic instruments and computer-based technology. He has many years of experience as a professional composer, musician, and music technologist, and has in these capacities been involved in numerous concerts, performance tours, and recordings across the USA and Europe.

This concert is supported by the UCD Michael Smurfit Graduate Business School and the UCD Institute for Discovery. CSMC is supported by Science Foundation Ireland.

%d bloggers like this:
search previous next tag category expand menu location phone mail time cart zoom edit close