Søren Tjagvad Madsen, Austrian Research Institute for Artificial Intelligence; and Gerhard Widmer, University of Linz
The paper presents a novel approach to exploring similarities in music performances. Based on simple measurements of timing and intensity in 12 recordings of a Schubert piano piece, short "performance archetypes" are calculated using a SOM algorithm and labelled with letters. Approximate string matching done by an evolutionary algorithm is applied to find similarities in the performances represented by these letters. We present a way of measuring each pianist's habit of playing similar phrases in similar ways and propose a ranking of the performers based on that. Finally, an experiment revealing common expression patterns is briefly described.