The 34 revised full papers presented were carefully reviewed and selected from participants' contributions and invited lectures given at the workshop. The volume is organized in two parts; the first on Interactive and Unsupervised Multimodal Systems contains 14 papers. The papers deal with the theoretical and computational issue of defining algorithms, programming languages, and determinist models to recognize
and synthesize multimodal signals. These are facial and vocal
expressions of emotions, tones of voice, gestures, eye contact, spatial arrangements, patterns of touch, expressive movements, writing patterns, and cultural differences, in anticipation of the implementation of intelligent avatars and interactive dialogue systems that could be exploited to improve user access to future telecommunication services. The second part of the volume, on Verbal and Nonverbal Communication Signals, presents 20 original studies devoted to the modeling of timing synchronisation between speech production, gestures, facial and head movements in human communicative expressions and on their mutual contribution for an effective communication.