Architecture for adaptive multimodal dialog systems based on voiceXML

This paper describes application oriented research on architectural building blocks and constraints for adaptive multimodal dialog systems that use VoiceXML as a component technology. The VoiceXML standard is well supported and promises to make the development of speech-enabled applications so easy that everyone with general web programming skills can accomplish it. The paper proposes a software architecture for multimodal interfaces that emphasizes modularity, in order to strengthen this potential by clearly separating different types of development tasks in a multimodal dialog system. The paper argues that adaptivity is a central design concern for multimodal dialog systems, but that adaptivity is not facilitated by the current VoiceXML standard. This and other examined limitations of VoiceXML for building multimodal dialog systems should be repaired in upcoming work on a successor standard that will explicitly target multimodal applications.