Abstract
This presentation covers a variety of research projects and contributions, ranging from sound analysis to music/movement interactions. The first part focuses on sound analysis and includes topics such as water sound detection and perception, urban soundscape visualization, and biodiversity assessment through ecoacoustics. Methods for automatic analysis range from audio descriptor identification to machine learning approaches. The second part of the presentation focuses on human synchronization. A review of human synchronization datasets helps us to frame this interdisciplinary concept. The presentation describes methods for capturing and evaluating synchronization between music and movement. These methods allow us to analyze various factors that influence synchronization, such as rhythm structure. Synchronization is also described through a biologically plausible model of the brain through spike neural networks. The final part of the presentation presents prototypes of sonification from human motion and brain activity.
Bio
Patrice Guyot is associate professor and musician. He received his master’s degree in Acoustics, Signal Processing, and Computer Science applied to Music (ATIAM) in 2010 at IRCAM (Paris). In 2014, he completed his Ph.D. in Informatics from the University of Toulouse for his work on water sound analysis. From 2014 to 2019, he worked at the Institut de Recherche en Informatique de Toulouse, University of Lille, and University of Sussex (UK), in different research projects related to sound perception, sound event detection, ecoacoustics and sound/movement interactions. Since 2019, Patrice Guyot is associate professor at IMT Mines Alès. Within the EuroMov Digital Health in Motion research unit, his research focus on digital audio tools and human movement. He currently coordinates the PIAS2 (Perception, Interaction, Synchronization and Signal) theme of the research unit, as well of the ModPuls project which focuses on synchronization modeling from spike neural networks. He is also co-chair of the forthcoming MOCO’26 (Motion + Computing) conference in Montpellier.