Laughter is a significant feature of human communication, and machines acting in roles like companions or tutors should not be blind to it. So far very limited progress has been made towards allowing computer-based applications to deal with laughter.ILHAIRE will lay the foundations of truly multimodal, multicultural laughter-enabled man-machine interaction, by associating experts of the different disciplines linked to this objective. The project developments will be assessed through the following measurable outcomes:1.A multicultural and multimodal database of laughter. This database will contain meta-data on the triggering conditions, and perceptual interpretation of it, annotated by psychologists.2.A classification of laughs depending on their types and meaning, as well as algorithms for audio-visual laughter analysis and classification3.Generative models of laughter based on statistical parametric synthesis algorithms; parameters will include results obtained from the abovementioned classification step.4.A characterization of temporal features of laughter and of the temporal relationship between multimodal signals (breathing and inhalation; motion of thorax, shoulder, etc.)5.A test-case man-machine dialogue system capable of laughing in real interactive situations. We will study how and when to respond to human laughter, as in a contagion loop, and when to trigger machine laughter.6.An understanding of the impact of such a laugh-enabled machine on users (benefits: loosening up the atmosphere, avoiding frustration; drawbacks: fear of being laughed at; humiliation; malicious laugh).This document describes these challenges as well as proposed methods to take them up. The project is also positioned in relation with the state-of-the-art and the benefits Ilhaire will bring to connected projects. Finally, the project participants, the overall management strategy and the vision to address the related issues (like IPR, ethics, etc.) are presented.