Educational and clinical techniques for language intervention in children with autistic spectrum disorders (ASD) focus on achieving a complete, speech-to-speech, communicative loop. To date, the AI technologies developed in areas like speech recognition, natural language processing, student modelling and intelligent tutoring have not been applied to the specific needs of children with ASD. In this paper we describe the design of Simone Says, a proposed software environment in which young children can practice semantically and socially meaningful language by playing a sort of interactive, linguistic game of Simon Says. Current research and practice in remediation both stress the need for achieving engagement and sustaining motivation in taking appropriate conversational turns and using language in functionally appropriate ways. Simone Says is intended to meet these requirements by using the natural attraction of computers to create opportunities for meaningful, speech-based language practice in a highly simplified social setting. In exercises that progress from vocabulary building to simple social conversation, the system will automatically generate contexts in which the student is rewarded for meaningful responses as defined by his or her current position along the normal developmental progression.

23 pages

Return to: SCS Technical Report Collection
School of Computer Science homepage

This page maintained by reports@cs.cmu.edu