Designing Visual Interfaces to Support Voice Input

Abstract

In the last years, voice has been used as an interaction mode in several devices such as smart speakers, mobile devices or television (TV). However, there seems to be a lack of practical recommendations about how visual interfaces can be designed to support the use of voice input in TV applications in certain tasks in a way to take advantage of the full potential of this interaction mode. This paper aims to present the design strategies adopted in a TV application targeting older adults which allows them to request help from a group of volunteers in daily life tasks and where voice is used in certain tasks such as to make the request instead of using the TV remote control. These strategies are discussed around three dimensions considered especially important when using voicebased input: information for voice usage, keeping the user informed and prompts to clarify system's doubts. The paper also presents the results of usability tests with potential users, in order to identify usability problems regarding the use of voice-based input to make the request and to gather information about how improve the usability of this feature. Study results indicate positive attitudes towards the use of voice input method and that older adults considered the use of voice easier than if they had to use a TV remote control to make a request. In a context where voice interaction appears as relevant for older adults, this study can contribute to more informed design strategies in multimodal interaction targeting this group. CCS CONCEPTS • Human-centered computing~User interface design • Humancentered computing~Natural language interfaces • Humancentered computing~Text input