A Polyphony Learning Software (PLS) based on gesture recognition by fuzzy inference system

Abstract

We are currently working on the realization of a PLS: Polyphony Learning Software tool within the framework of a pluridisciplinary project for promoting and preserving the Corsican culture. In this article we shall describe the different technical choices adopted so as to develop this learning tool based upon a gestural language. This software should enable a single singer to create polyphony. This is possible by replacing the other singers with singing voices. Our problematic is to associate artificial voices to each of the language gestures. This process has been broken down into several steps: defining a gestural language, gesture capture and recognition, association of a singing voice.