The workshop “Methods and Tools for Audio-Visual Processing and Human Robot Interaction” will be held at Inria (Montbonnot, Room F107), on Tuesday, August 23rd, 2022. We have the pleasure to host two wonderful speakers: Prof. Gaël Richard and Prof. Simon Leglaive. Completed and ongoing works of the members of the chair will also be presented.

Invited talks:

[9h30] Unsupervised Audio Source Separation Using Differentiable Parametric Source Models by Prof. Gaël Richard

[10h15] Learning and controlling the source-filter representation of speech with a variational autoencoder by Prof. Simon Leglaive     [slides]

Talks from the chair members:

[10h40] An online Minorization-Maximization algorithm by Dr. Florence Forbes

[11h05] Face processing for visual- and audio-visual speech by Dr. Radu Horaud    [slides]

[11h30] An introduction do dynamical variational autoencoders by Dr. Xavier Alameda-Pineda     [slides]

Lunch Break

[14h00] Dynamic neural fields and manifold learning for audiovisual fusion in psychophysics and robotics by Simon Forest     [slides]

[14h25] Deep Transfer Reinforcement Learning for Social Robotics by Dr. Chris Reinke

[14h50] Open discussion

Comments are closed.