Interactive Computation of Timbre Spaces for Sound Synthesis Control

Abstract

Expressive sonic interaction with sound synthesizers requires the control of a continuous and high dimensional space. Further, the relationship between synthesis variables and timbre of the generated sound is typically complex or unknown to users. In previous works, we presented an unsupervised mapping method based on machine listening and machine learning techniques, which addresses these challenges by providing a low-dimensional and perceptually related timbre control space. The mapping maximizes the breadth of the explorable sonic space covered by the sound synthesizer, and minimizes possible timbre losses due to the low­-dimensional control. The mapping is generated automatically by a system requiring little input from users. In this paper we present an improved method and an optimized implementation that drastically reduce the time for timbre analysis and mapping computation. Here we introduce the use of the extreme learning machines for the regression from control to timbre spaces, and an interactive approach for the analysis of the synthesizer sonic response, performed as users explore the parameters of the instrument. This work is implemented in a generic and open-source tool that enables the computation of ad hoc synthesis mappings through timbre spaces, facilitating and speeding up the workflow to get a customized sonic control system.