Robust and Adaptive Architecture for Multilingual Spoken Dialogue Systems

We present how robustness and adaptivity can be supported by the spoken dialogue system architecture. AthosMail is a multilingual spoken dialogue system for e-mail domain. It is being developed in the EU-funded DUMAS project. It has flexible system architecture supporting multiple components for input interpretation, dialogue management and output generation. In addition to language differences, these components have great variation in their approaches for spoken interaction. For example, the dialogue management components implement different dialogue control models. The system architecture makes the interaction robust and adaptive by utilizing different approaches for spoken interaction in a single application by selecting suitable components at runtime.