Compansion: From research prototype to practical integration

Abstract

Augmentative and Alternative Communication (AAC) is the field of study concerned with providing devices and techniques to augment the communicative ability of a person whose disability makes it difficult to speak or otherwise communicate in an understandable fashion. For several years, we have been applying natural language processing techniques to the field of AAC to develop intelligent communication aids that attempt to provide linguistically correct output while increasing communication rate. Previous effort has resulted in a research prototype called Compansion that expands telegraphic input. In this paper we describe that research prototype and introduce the Intelligent Parser Generator (IPG). IPG is intended to be a practical embodiment of the research prototype aimed at a group of users who have cognitive impairments that affect their linguistic ability. We describe both the theoretical underpinnings of Compansion and the practical considerations in developing a usable system for this population of users.

Personal storytelling: Using Natural Language Generation for children with complex communication needs, in the wild…

“…In their review of Natural Language Processing (NLP) and AAC, [12] suggest that nlg could be used to generate complete utterances from the limited input that AAC users are able to provide. For example, the Compansion project [13] used nlg techniques to expand telegraphic user input, such as "Mary go store? ", into complete utterances, such as "Did Mary go to the store?".…”

“…11 http://nyxdevices.com/product/, retrieved April 2015 12 http://www.fitbit.com/, retrieved April 2015 13 We used the QR 2D barcode, http://www.qrcode.com/en/, retrieved November 2015 13 • Location data -each time the child changes location, in particular when they go somewhere unexpected (i.e., to a location not specified in the timetable, such as the kitchen during class).…”

A semantic grammar for beginning communicators

“…From the specific point of view of AAC, there are several efforts to develop strategies to improve the communication rates in order to improve the comprehension or the acquisition of communicative skills of early beginners. One of the first proposals was the COMPANSION project [25] which used NLP techniques to expand telegraphic user input, such as ''store Mary go? '', into complete utterances, such as ''Did Mary go to the store?''.…”

Towards the improvement of Augmentative and Alternative Communication through the modelling of conversation

“…A number of NLP techniques have been investigated in AAC Newell et al, 1998). The Compansion system was designed to process and expand spontaneous language constructions such that a group of uninflected content words could be converted automatically into a fully formed phrase or sentence (Demasco and McCoy, 1992;McCoy et al, 1998), for example, and syntax processing was investigated for use within predictive AAC interfaces (Swiffin et al, 1987a;Morris et al, 1991;Yang et al, 1990).…”