Based on the framework of partially observable Markov decision processes (POMDPs), this paper describes a practical real-time spoken dialogue system in which the underlying belief state is represented by a dynamic Bayesian Network and the policy is parameterized using a set of action-dependent basis functions. Tractable real-time Bayesian belief updating is made possible using a novel form of Loopy Belief Propagation and policy optimisation is performed using an episodic Natural Actor Critic algorithm. Details of these algorithms are provided along with evaluations of their accuracy and efficiency. The proposed POMDP-based architecture was tested using both simulations and a user trial. Both indicated that the incorporation of Bayesian belief updating significantly increases robustness to noise compared to traditional dialogue state estimation approaches. Furthermore, policy learning worked effectively and the learned policy outperformed all others on simulations. In user trials the learned policy was also competitive, although its optimality was less conclusive. Overall, the Bayesian update of dialogue state framework was shown to be a feasible and effective approach to building real-world POMDP-based dialogue systems.

Although user simulations are increasingly employed in the development and assessment of spoken dialog systems, there is no accepted method for evaluating user simulations. In this paper, we propose a novel quality measure for user simulations. We view a user simulation as a predictor of the performance of a dialog system, where per-dialog performance is measured with a domain-specific scoring function. The quality of the user simulation is measured as the divergence between the distribution of scores in real dialogs and simulated dialogs, and we argue that the Cramér-von Mises divergence is well-suited to this task. The technique is demonstrated on a corpus of real calls, and we present a table of critical values for practitioners to interpret the statistical significance of comparisons between user simulations. Index Terms — User simulation, user modelling, dialog simulation, dialog management

fission output, natural language generation, visual-language generation. In recent years, we have developed a framework of humancomputer interaction that offers recognition of various communication modalities including speech, lip movement, facial expression, handwriting and drawing, body gesture, text and visual symbols. The framework allows the rapid construction of a multimodal, multi-devices, and multi-user communication system within crisis management. This paper reports the multimodal information presentation module combining language, speech, visual-language and graphics, which can be used in isolation, but also as part of the framework. It provides a communication channel between the system and users with different communication devices. The module is able to specify and produce context-sensitive and user-tailored output. By the employment of ontology, it receives the system’s view about the world and dialogue actions from a dialogue manager and generates appropriate multimodal responses.