Dialogue Interaction with the DARPA Communicator Infrastructure: The Development of Useful Software

Dialogue Interaction with the DARPA Communicator Infrastructure: The Development of Useful Software Samuel Bayer The MITRE Corporation 202 Burlington Rd. Bedford, MA 01730 sam@mitre.org Christine Doran The MITRE Corporation 202 Burlington Rd. Bedford, MA 01730 cdoran@mitre.org Bryan George The MITRE Corporation 11493 Sunset Hills Rd. Reston, VA 20190 bgeorge@mitre.org ABSTRACT To support engaging human users in robust, mixed-initiative speech dialogue interactions which reach beyond current capabilities in dialogue systems, the DARPA Communicator program [1] is funding the development of a distributed message-passing infrastructure for dialogue systems which all Communicator participants are using. In this presentation, we describe the features of and requirements for a genuinely useful software infrastructure for this purpose. Keywords Spoken dialogue, speech interfaces 1. INTRODUCTION Over the last five years, three technological advances have cooperated to push speech-enabled dialogue systems back into the limelight: the availability of robust real-time speech recognition tools, the explosion of Internet-accessible information sources, and the proliferation of mobile information access devices such as cell phones. However, the systems being fielded, and the standards arising from these efforts, represent only a limited set of capabilities for robust voice-enabled interaction with knowledge sources. The most prominent indication of these limitations is the fact that these systems are overwhelmingly system-directed; the system asks a question, and the user responds. While this type of interactions sidesteps a number of problems in speech recognition and dialogue tracking, it is overwhelmingly likely that these restrictions are not manageable in the long term. The DARPA Communicator program [1] is exploring how to engage human users in robust, mixed-initiative speech dialogue interactions which reach beyond current capabilities in dialogue systems. To support this exploration, the Communicator program has funded the development of a distributed message-passing infrastructure for dialogue systems which all Communicator participants are using. In this presentation, we describe the features of and requirements for a genuinely useful software infrastructure for this purpose. 2. BUILDING USEFUL SOFTWARE The Galaxy Communicator software infrastructure (GCSI) is an elaboration and extension of MIT's Galaxy-II distributed infrastructure for dialogue interaction [3]. The fact that all program participants are required to use the GCSI imposes a somewhat more severe set of requirements on the infrastructure than usual, and these requirements range far beyond the straightforward considerations of functionality. • Flexibility: the infrastructure should be flexible enough to encompass the range of interaction strategies that the various Communicator sites might experiment with • Obtainability: the infrastructure should be easy to get and to install • Learnability: the infrastructure should be easy to learn to use • Embeddability: the infrastructure should be easy to embed into other software programs • Maintenance: the infrastructure should be supported and maintained for the Communicator program • Leverage: the infrastructure should support longer-term program and research goals for distributed dialogue systems 3. FLEXIBILITY The GCSI is a distributed hub-and-spoke architecture based on message-passing. The hub of the GCSI incorporates a scripting mechanism that allows the programmer to take control of the message traffic by implementing "hub programs" in a simple …

Content truncated to 500 words