Elementos estruturantes para o projeto de interfaces multimodais

Abstract

Interfaces tend increasingly to become multimodal. Which is the combination of multiple input and output information resources that allow user interaction with software, enabling the choice of the desired mode, (e.g. voice, gestures and touch), and to switch between them. Digital artifacts available on the web can also be reconstructed according to different modes, enabling a new interpretation and interaction with them. The development of solutions according to this perspective has shown promise in increasing the accessibility to people with visual impairments, despite that they often present new challenges to that category of users. For instance, this happens with touchscreen interfaces and sites with dynamic features, such as animations. This thesis proposes a multimodal interaction framework to support people with visual impairment. Two views of this structure have been developed: the first, RotaColab, is intended for navigational guidance using mobile devices and the second, WABlind, is intended for the reconstruction of web content in order to enable the use of screen readers. Proofs of concept and application of Semiotics Engineering methods were used to evaluate the two developed prototypes, demonstrating the feasibility of the proposed structure.