Spoken language is a natural way to control the human-robot interaction, especially for mobile service robots. It has some important advantages over other communication approaches: eyes and hands free, communication from a distance, even without being in line of sight and no need for additional learning for humans. In this paper, we present the spoken dialog framework integrated in our mobile service robot CAMBADA@Home, a robotic platform aimed at move into a living space and interact with users of that space. The proposed framework comprises three major spoken and natural language processing components: an Automatic Speech Recognition component to process the human requests, a Text-to-Speech component to generate more natural responses from the robot side, and a dialog manager to control how these two components work together.