In the context of audio and video document indexing and retrieval, the main studies of the team deal with the correlated extraction of the audio and video information and its exploitation to propose intelligent access. This research implies studies on signal processing, on modeling and learning and also on audiovisual content structuring.