AN AI DRESS REHEARSAL: EXPLORING MUSIC PERFORMANCE AND INTERACTION WITH COMPUTATIONAL MODELS
Carlos Cancino-Chacón (Institute of Computational Perception at Johannes Kepler University Linz, AT)

The way a piece of music is performed is a very important factor influencing our enjoyment of music.  A good performance goes beyond a precise rendering of the score; performers shape aspects like tempo, dynamics, and articulation to convey emotion and engage listeners.

This talk focuses on a specific area of research: computational models of expressive music performance. These models aim to codify hypotheses about expressive performance using mathematical formulas or computer programs, enabling systematic and quantitative analysis. The models serve two main purposes: they allow us to systematically test hypotheses about how music is performed, and they can be used as tools to create automated or semi-automated performances in artistic and educational settings.

In this talk, I will explore two key aspects: data-driven approaches to modeling expressive performance and interdisciplinary collaboration with music cognition to understand how humans interact and develop expressive interpretations. I will illustrate these aspects through three main topics: (1) Basis Function Models, a machine learning framework for generating expressive performances based on musical scores; (2) Studying human interaction in musical performance and insights into the development of a real-time automatic accompaniment system; and (3) The Rach3 Project, an investigation into how pianists learn new music and develop their own expressive interpretations.

Carlos Cancino-Chacón is a Senior Scientist at the Institute of Computational Perception at Johannes Kepler University Linz (JKU), Austria, and the Principal Investigator of the Rach3 and AURA projects, both funded by the Austrian Science Fund. He previously conducted research at the Austrian Research Institute for Artificial Intelligence (OFAI) and was a Guest Researcher at the RITMO Centre, University of Oslo. His research focuses on machine learning models for understanding music performance and listening, with an emphasis on three areas: computational modeling of expressive performance, (real-time) human–computer interaction in music, and cognitively plausible machine listening.

In English

Add event to my calendar

Idea & Organization:
Simon Blatt, InterMediation / Mathematics Department, University of Salzburg
Katarzyna Grebosz-Haring, InterMediation /Inter-University Organization Science & Arts, Mozarteum University Salzburg, University of Salzburg

Organized by focus area (Inter)Mediation. Music – Mediation – Context / Inter-University Organization Science & Arts, University of Salzburg/Mozarteum University Salzburg in cooperation with the Mathematics department, University of Salzburg in series Music, the Arts & Math.