We present QUARK (QUestion Answering Rendering Knowledge), an
architecture and a prototypical first partial implementation to answer customer
questions in the telecommunication domain from several knowledge sources. The system features a central dialog manager instance that collects answers for a given question from several knowledge sources and selects the most probable based on a combination of confidence value and a weight for each knowledge source “per se”.
Felix Burkhardt does tutoring, consulting, research and development in the working fields human-machine dialog systems, text-to-speech synthesis, speaker classification, ontology based natural language modeling, voice search and emotional human-machine interfaces.
Originally an expert of Speech Synthesis at the Technical University of Berlin, he wrote his ph.d. thesis on the simulation of emotional speech by machines, recorded the Berlin acted emotions database, "EmoDB", and maintains several open source projects, including the emotional speech synthesizer "Emofilt" and the speech labeling and annotation tool "Speechalyzer". He has been working for the Deutsche Telekom AG since 2000, currently for the Telekom Innovation Laboratories in Berlin.
