Tristan Jehan

Skeleton is a Toolkit for the Analysis,
Modeling, and Synthesis of Music



Skeleton is a set of Cocoa libraries optimized for fast development of research, and professional applications dealing with the analysis of musical signals. Based on fundamentals of psychoacoustics, perception, and learning, the underlying framework (see figure 1) consists of machine listening, and machine learning tools, supported by flexible data structures, and visualizations.


Figure 1: Framework of the perceptual analysis, and learning model


Designed essentially to support my research as an alternative to more generic and slower tools such as Matlab, Skeleton should provide a robust, efficient, and specific, yet open programming environment for deconstructing, structuring and labeling audio, or for generating personalized music (see figure 2).


Figure 2: Personalized Music Synthesizer


Skeleton aims at combining various solid scientific approaches to music listening, learning, and cognition, to enable the development of consistent creative audio applications, and enhance the music making.


Block diagram of the environment
(click for full size image)