There is an increasing need for mobile spoken dialogue systems. Mobile devices, such as smartphonesand personal digital assistants, can be used to implement efficient speech-based and multimodal interfaces. Currently, the development of such applications lacks suitable tools. We introduce general system architecture for mobile spoken dialogue systems. The architecture is a mobile version of the existing Jaspis architecture. The architecture makes it possible to implement distributed dialoguesystems using compact software agents in flexible ways both on the server and mobile devices. We present how an existing server-based timetable application is turned into a multimodal mobile application. The server handles the overall coordination of the dialogue by generating dialogue task descriptions using VoiceXML, and the mobile device realizes the dialogue descriptions with available resources.
Cite as: Turunen, M., Salonen, E.-P., Hakulinen, J., Kanner, J., Kainulainen, A. (2005) Mobile architecture for distributed multimodal dialogues. Proc. Applied Spoken Language Interaction in Distributed Environments (ASIDE 2005), paper 22
@inproceedings{turunen05_aside,
author={Markku Turunen and Esa-Pekka Salonen and Jaakko Hakulinen and Juuso Kanner and Anssi Kainulainen},
title={{Mobile architecture for distributed multimodal dialogues}},
year=2005,
booktitle={Proc. Applied Spoken Language Interaction in Distributed Environments (ASIDE 2005)},
pages={paper 22}
}