Linguistic Knowledge Sources for Spoken Language Understanding

The objective of the Unisys Spoken Language Systems effort is to develop and demonstra te technology for the understanding of goal-directed spontaneous speech. The Unisys spoken language architecture couples speech recognition systems with the Unisys discourse understanding system, PUNDIT. PUNDIT is a broad-coverage language understanding system used in a variety of message understanding applications and has been extended to handle spoken input. I ts power comes from the integrat ion of syntax, semantics, and pragmat ics processing, the abil i ty to por t rapidly to new task donudns, and from an open, modula,r a~chitecture. PUNDIT is unique in i ts abil i ty to handle connected discourse; i t includes a reference resolution module tha t tracks discom~se entit ies and distinguishes references to previously mentioned entit ies from the introduct ion of new entities. The PUNDIT front-end supports tunttaking dialogue and permits the system to include bo th questions and answers in hnilding an in tegrated discourse context, required for the handling of interactive communication. PUNDIT has been interfaced to several speech recognition systems ( M I T SUMMIT and I T T V R S 1 2 8 0 ) to perform applications in dixection-finding assistance, a l l travel planning and air traffic control.