This workshop is intended to introduce the CMMR community to the use of data science tools (network theory, machine learning etc.) in the field of music theory, analysis and composition, and the physical and perceptual characterization of sound. The format will be mixed, with lecture-style and hands-on sections. Participants will be introduced to the MUSICNTWRK package (www.musicntwrk.com), a python library comprised of four modules:
Participants will learn how to work with standard data manipulation and visualization tools, like Pandas, TensorFlow, Gephi, etc. and how to integrate all of the above in a compositional or analytic framework. Previous knowledge of Python and familiarity with programming languages is recommended. Participants will need to have their own laptop to take full advantage of the hands-on sections. All exercises will be run using the Google Colaboratory cloud environment and jupyter notebooks, so there will be no restrictions on individual devices, operating systems or software installations.
Big data tools have become pervasive in virtually every aspects of culture and society. In music, application of such techniques in Music Information Retrieval applications are common and well documented. However, a full approach to musical analysis and composition is not yet available for the music community and there is a need for providing a more general education on the potential, and the limitations, of such approaches. From a more fundamental point of view, the abstraction of musical structures (notes, melodies, chords, harmonic or rhythmic progressions, timbre, etc.) as mathematical objects in a geometrical space is one of the great accomplishments of contemporary music theory. Building on this foundation, the organizer has generalized the concept of musical spaces as networks and derived functional principles of compositional design by the direct analysis of the network topology. This approach provides a novel framework for the analysis and quantification of similarity of musical objects and structures, and suggests a way to relate such measures to the human perception of different musical entities. Finally, the analysis of a single work or a corpus of compositions as complex networks provides alternative ways of interpreting the compositional process of a composer by quantifying emergent behaviors with well-established statistical mechanics techniques. Interpreting the latter as probabilistic randomness in the network, the organizer developed novel compositional design frameworks that are central to his own artistic research.
This workshop is intended to provide CMMR participant with a hands-on introduction to data tools as implemented in the MUSICNTWRK code developed by the organizer and freely available at www.musicntwrk.com. Combining lecture and tutorial sections, the participants will be able to explore the main features of the MUSICNTWRK combined with other popular data management and visualization tools (Pandas, TensorFlow, Gephi, etc.). The goal is to provide enough information that the participant will be able to integrate this framework within their own scientific or artistic practice.
Marco Buongiorno Nardelli
E-mail: mbn (at) unt.edu
Marco Buongiorno Nardelli is University Distinguished Research Professor at the University of North Texas: composer, media artist, flutist, computational materials physicist, and a member of CEMI, the Center for Experimental Music and Intermedia, and iARTA, the Initiative for Advanced Research in Technology and the Arts. He is a Fellow of the American Physical Society and of the Institute of Physics, and a Parma Recordings artist.
Date: Friday 18 October (full day)
Capacity: 10 for hands-on activity + 10 for auditing
Location: Salle de conférences Pierre Desnuelle, CNRS Campus Joseph Aiguier, 31 chemin Joseph Aiguier, 13009 Marseille