Multimodal Artefact Manipulation: Evaluation in Real Contexts

The evaluation of a mobile application must be done considering its purpose and contexts on which it is suppose to be used. In this paper we introduce a multimodal artefact manipulation tool and discuss the results of its evaluation: conducted with a sub-set of its final users, in real contexts, considering: different interaction modalities (voice recognition, gesture recognition, direct interaction, and keypad), and context variables (device, noise, lighting and movement status). This evaluation enabled us to understand: user acceptance of the tool; and, the impact of specific context variables on the performance of the different modalities.