Verb Prediction and the Use of Particles for Generating Sentences in AAC System
This study proposes and develops a novel Augmentative & Alternative Communication (AAC) system which improves the verbal communication of speech disordered people. The success of AAC systems is largely determined by two factors: the selection of vocabularies appropriate to the occasions and the efficient utilization of limited display space in AAC systems. To meet the two vital requirements, vocabulary databases are built and embedded into our AAC system, thus the context-aware predicate prediction and the particle use in generated sentences even in a limited display space are feasible. We construct the databases by collecting utterances in everyday life dialogues by recording, sorting frequently used vocabularies, and applying semantic and lexical interpretation.
KeywordsSelection Limitation Speech Disorder Correct Communication Sentence Generation Situation Information
Unable to display preview. Download preview PDF.
- 1.Glennen, S.L., De Cote, D. (eds.): The Handbook of Augmentative and Alternative Communication. Singular Publishing Group, Inc. (1996)Google Scholar
- 3.McCoy, K.F., Demasco, P.: Some Applications of Natural Language Processing to the Field of Augmentative and Alternative Communication. In: Proceedings of the IJCAI 1995 Workshop on Developing AI Applications for People with Disabilities, Montreal, Canada (August 1995)Google Scholar
- 4.O’Donnell, M.: Input specification in the WAG sentence generation system. INLG 8, 41–50 (1996)Google Scholar