An architecture for designing distributed spoken dialogue interfaces

A spoken dialogue interface allows a user to interact with a computer application using speech. The user engages in a conversation with the application to achieve some goal, for example to obtain travel information or to book theatre tickets. In this paper, we describe an architecture and development environment for designing distributed spoken dialogue interfaces. A distributed spoken dialogue interface allows multiple users, distributed throughout a computer network, to interact with an application using speech. With a distributed spoken dialogue interface, several users engage in a conversation with the application, at different times, to achieve some goal. Our approach to distributed spoken dialogue interfaces is based on the idea of an intelligent agent that coordinates the activities of the multiple users interacting with the application. To support our model of distributed spoken dialogues, we have created a software development environment, JSBB, that can be used to design both distributed and non-distributed spoken dialogue interfaces.