الفهرس | Only 14 pages are availabe for public view |
Abstract An interactive dialogue management system for spoken dialogue system controls the overall interaction between the system and the user, and coordinates activities of all spoken dialogue system’s components. Spoken dialogue system is one of the most important human-machine communication ways. Dialogue system seeks to provide a natural conversational interaction between the user and the computer system in an intelligent behavior. The spoken dialogue system consists of many phases; the most important one is the dialogue management, which is the core of the system, so the main target of this research is to build an interactive dialogue management system for spoken dialogue system in ideal way. The ideal way means that, the spoken dialogue system is to be less complexity and use uncertainty model; this way must be acceptable by user and the system itself.As a trial to get the optimal solution, this thesis presents a proposed model for the management process, this model uses a Makov Decision Process (MDP), the MDP supposes that the surrounded environment is ideal. So; this research operates on all phases of the spoken dialogue system, theses phases are effect on the dialogue management. So the main tasks of this research were accomplished in the following steps: •Automatic speech recognition system was implemented to recognize the speech of the user. This system receive a sentence and segment it into words, then recognize these words, in order to improve the performance, Principal component analysis was used to reduce the feature vector. •A proposed system which gathers the natural language understanding and dialogue management was implemented. •A natural language generation system was established; this system used A Hidden Markov Model. •Text to speech system was built, and to get better results a comparison study was done between different signal smoothing techniques in order to get a sound in a natural way. |