On recent years there has been increasing emphasis on research for speech and video processing for the meeting room domain. Within the different projects that are interested in this area, two different alternative meeting settings have been proposed. On one hand, some consider a lecture environment where a single speaker gives a talk in front of an audience, which intervenes at different points of the lecture with questions and remarks. In this situation there is always a main speaker, facing an audience, and many people listening, facing the speaker. On the other hand, the conference room environment is a gathering of people where mostly everyone speaks and discussions are being carried on one or more common topics to all attendees.