This book constitutes the thoroughly refereed post-proceedings of the 4th International Workshop on Machine Learning for Multimodal Interaction, MLMI 2007, held in Brno, Czech Republic, in June 2007. The 25 revised full papers presented together with 1 invited paper were carefully selected during two rounds of reviewing and revision from 60 workshop presentations. The papers are organized in topical sections on multimodal processing, HCI, user studies and applications, image and video processing, discourse and dialogue processing, speech and audio processing, as well as the PASCAL speech separation challenge.