Constructing Sonic Structures with Visual Roboser

Márcio de Oliveira Costa
Jonatas Manzolli

Núcleo Interdisciplinar de Comunicação Sonora
Universidade Estadual de Campinas (Unicamp)
Caixa Postal 6166 ­ 13091 - 970
Campinas ­ SP ­ Brazil

Paul Verschure

Institute of Neuroinformatics - INI, ETHZ


This paper describes the basic system architecture of Visual Roboser that transforms image sequences into sonic structures. It is an extension of the Roboser Project and it uses images captured from the environment in real time to drive the production of sonic structures. It is based on an architecture that integrates a neural simulator and two JAVA modules: one for image processing and another for sound generation. Possible applications of Visual Roboser include real time composition, multimedia live performances, sonic installations and intelligent interactive rooms such as our latest work name "Ada: the Intelligent Space" (

Download paper: [pdf] 194684 bytes