Mathematical morphology for the analysis and generation of time-frequency representations of music
Gonzalo Romero-García, a doctoral student at Sorbonne University, completed his thesis entitled "Mathematical morphology for the analysis and generation of time-frequency representations of music" in the STMS laboratory (Ircam - Sorbonne University - CNRS - Ministry of Culture). He has been awarded a doctoral contract by EDITE (ED130).
He will defend his thesis at Ircam on 15 November 2023 at 2pm, in English, before a jury comprising :
- Florent Jacquemard, Researcher at Inria - Reviewer
- Laurent Najman, Professor at Université Gustave Eiffel - Reviewer
- Thierry Géraud, Professor at Epita - Examiner
- Julien Tierny, Director of research at CNRS - Examiner
- Dmitri Tymoczko, Professor at Princeton University - Examiner
- Isabelle Bloch, Professor, Sorbonne Université - Thesis Co-director
- Carlos Agon, Professor, Sorbonne Université - Thesis Co-director
The presentation will be in English, at Ircam, but can also be followed live on Ircam's YouTube channel: https://youtube.com/live/GNKjLbUzLGk
This thesis explores the application of Mathematical Morphology to music analysis and generation, focusing on two time-frequency representations: spectrograms and piano rolls. Mathematical morphology is a non-linear image processing tool used to consider topological notions of the image. We present three applications. The first is to analyse spectrograms using morphological tools to obtain parameters with which to synthesise the sound of a musical instrument. The second is to generate piano rolls with two musical parameters, texture and harmony, by arranging them using morphological dilation. The third is to apply morphological operators to analyse piano rolls using graph theory. The thesis thus proposes new approaches to problems in sound analysis and computational musicology.