Creating audio solutions for robots is no easy task! Two of the leading experts in this field will talk us through the hardware limitations, algorithm development and challenging environments. How far have we come, and what still needs to be solved? Is it even possible to get there with the technology that is available today?
Program
Francois and Caleb will take us through everything to do with robot audition in a joint presentation.
We will begin with an introduction to the subject by describing the auditory scene, tasks and characteristics of robot audition. Then we will take a deeper look at signal processing methods, model-based methods (end-to-end deep learning) and the pros and cons of each.
We will hear about the efforts by both universities: Université de Sherbrooke (Canada) and UNAM (Mexico) in localization, separation and classification.
To finish off we’ll have a discussion about the research and development, and also other use cases that can benefit, such as bioacoustics, hearing-aid research and urban/domestic sound design.
Francois Grondin, Université de Sherbrooke
Born and raised in Montréal (Québec), François Grondin received his bachelor’s degree (’09) in Electrical Engineering from McGill University, and his Master’s degree (’11) and Ph.D. degree (’17) in Electrical Engineering from the Université de Sherbrooke. He then worked as a Postdoctoral Associate at Massachusetts Institute of Technology in CSAIL for two years, and he is currently an assistant professor in the Department of Electrical Engineering and Computer Engineering at Université de Sherbrooke since 2020. He is a member of IntRoLab at the Interdisciplinary Institute for Technological Innovation in Sherbrooke, where his research focuses mainly on robot audition.
Caleb Rascon, UNAM
Caleb Rascon founded Makko IT, a company dedicated to the development of digital technology, in which he is now a passive partner, and advisor on topics such as Digital Signal Processing and Artificial Intelligence. He obtained his PhD at the University of Manchester (UK, 2009). He was awarded, along with the Golem Group, the Innovation Award of RoboCup@Home (Netherlands, 2013) for his efforts in Robot Audition. He was also awarded the distinction of Innovator under 35 by MIT Technology Review (Mexico, 2014). He is currently a Research Associate at the Universidad Nacional Autonoma de Mexico (UNAM) conducting research in the area of Auditory Scene Analysis.
If you have questions about signing up, please contact Murielle De Smedt, mds@danishsound.org