Robust Dialog Management Architecture Using VoiceXML for Car Telematics Systems

Abstract

12This paper describes a dialog management architecture for car telematics systems. The system handles the user's spontaneous utterances and the variable communication conditions between the in-car client and the server. The communication is based on VoiceXML/HTTP, and the development of the server-side application is based on layered extensions of VoiceXML, called DialogXML and ScenarioXML. These extensions provide support for state-and-transition type programming, enable access to dynamic external databases, and share commonly used dialogs via templates. The client system includes a set of small grammars and lexicons for various tasks; only relevant grammars and lexicons are activated under the control of the dialog manager. The server-side applications are integrated via an abstract interface, and the client system may include compact versions of the same applications. The VoiceXML interpreter can switch between applications on both sides intelligently. This helps to reduce bandwidth utilization, and allows the system to continue even if the communication channel is lost.

VoiceXML for Pervasive Environments

“…A mixed-initiative dialog strategy is realized by the integration of a natural language analyzer which produces slot/filler pairs from a given user utterance for the current dialog state. They subsequently employed their system to implement a voice user interface for a car (Obuchi et al, 2005).…”

Robust speech dialog interface for car telematics service

“…Second, we discuss the creation of a speech dialog interface for HMIs, and report on a new dialog management system called CAMMIA [3] [4]. Speech Dialog is an essential function to realize sophisticated and human-centered HMIs.…”