Supporting Mobile Multimodal Interaction with a Rule-Based Framework

Abstract: Multimodality can make (especially mobile) device interaction more efficient. Sensors and communication capabilities of modern smartphones and tablets lay the technical basis for its implementation. Still, mobile platforms do not make multimodal interaction support trivial. Building multimodal applications requires various APIs with different paradigms, high-level interpretation of contextual data, and a method for fusing individual inputs and outputs. To reduce this effort, we created a framework that simplifies and accelerates the creation of multimodal applications for prototyping and research. It provides an abstraction of information representations in different modalities, unifies access to implicit and explicit information, and wires together the logic behind context-sensitive modality switches. In the paper, we present the structure and features of our framework, and validate it by four implemented demonstrations of different complexity.