Towards a Reference Architecture for Augmentative and Alternative Communication Systems

Abstract

Augmentative and Alternative Communication (AAC) systems are broadly used to expand communication skills of individuals with severe communicative disorders. These systems should prioritize requirements that contribute to making the communication easy and effective to be robust. However, there is no guideline available to aid the implementation of these systems. In this paper, we analyze the fragmented knowledge available on scientific proposals and technological solutions to extract key concepts for a robust AAC system. Then, we propose a Reference Architecture (RA) that can be used as a conceptual guide to define and evaluate concrete and robust software architectures for several goals including literacy in inclusive education.

Requesting Help Module Interface Design on Key Partial Video with Action and Augmented Reality for Children with Autism Spectrum Disorder

“…Researchers used different colors to classify the different parts of speech (for example, yellow for subject words, orange for affirmative and negative words, green for verbs, and so on). They also used a structured layout design to guide participants in using lexical phrases in proper sequence [42][43][44].…”

Pictogram Prediction in Alternative Communication Boards: a Mapping Study

“…An example of difficulty is the time needed to construct and communicate sentences [Pereira et al 2019, Berenguer et al 2022, making communication less fluid and causing frustration in the user and his interlocutor. To overcome this problem, a robust AAC system must support sentence construction by facilitating pictogram searching and selection [Franco et al 2018]. This requirement can be satisfied by using colors to classify pictograms by their part of speech (e.g., nouns, verbs, and adjectives) or by using prediction techniques to suggest the most appropriate next pictogram [Franco et al 2018].…”

“…To overcome this problem, a robust AAC system must support sentence construction by facilitating pictogram searching and selection [Franco et al 2018]. This requirement can be satisfied by using colors to classify pictograms by their part of speech (e.g., nouns, verbs, and adjectives) or by using prediction techniques to suggest the most appropriate next pictogram [Franco et al 2018].…”