Donald E. Walker, William H. Paxton, B. Grosz, Gary G. Hendrix, Ann E. Robinson, Jane J. Robinson, & J. Slocum
This paper describes the procedures for integrating knowledge from different sources in the SRI speech understanding system. A language definition system coordinates–at the phrase level–information from syntax, semantics, and discourse in the course of the interpretation of an utterance.
Ready Schools is a radical systems change effort to improve the academic, health, and social outcomes of historically underserved children in one of this country’s largest and most diverse communities, Miami-Dade County. This report presents Year 3 findings of SRI's evaluation.
Since receiving support from the W.K. Kellogg Foundation in 2007, Ready Schools partners have made significant progress toward improving the systems in Miami-Dade that contribute to school readiness and success. This report presents findings of the second year of evaluation.
To have a general-purpose machine vision capability, we must be able to recognize things; we argue that most natural objects have a part structure that we can recover from image data and thus use as the basis for "general-purpose" recognition.
Current shape-from-shading and shape-from-texture methods are applicable only to smooth surfaces, while real surfaces are often rough and crumpled. To extend such methods to real surfaces we must have a model that also applies to rough surfaces.
This research outlines the predominant dialogue and performance characteristics of three-person interpreted telephone speech during service-oriented dialogues, in comparison with those of two-person non-interpreted dialogues.
Two important problems in speech understanding are how to effectively integrate multiple sources of knowledge within the system and how to control the activities of the system to arrive at appropriate interpretations for utterances.
Fernando C. N. Pereira, Douglas E. Appelt, & Paul Martin
This paper describes the design of a transportable natural language (NL) interface to databases and the constraints that transportability places on each components of such a system.
We describe an approach to abductive reasoning called weighted abduction, which uses inference weights to compare competing explanations for observed behavior. We present an algorithm for computing a weighted-abductive explanation, and sketch a model-theoretic semantics for weighted abduction.
