The sound of fairness: An integrative framework for embedding fairness into voice interaction design
Year: 2025
Type: article
Abstract: Voice assistants (VAs) substantially affect the manner in which users interact with technology by offering convenient, hands-free user interfaces. However, owing to inherent limitations in conveying and visualising information, voice interactions are characterised by brief outputs that may, justifiably or not, trigger perceptions of unfairness among users. Although studies have developed general voice user interface (VUI) design guidelines to optimise VA-user interactions, they lack a focus on fairness. Neglecting fairness considerations can result in user dissatisfaction, a loss of trust, and lower service quality. To address this issue, we first assessed fairness in VUI design through 15 semi-structured expert interviews, linking four established fairness dimensions to VUI design guidelines and decomposing each dimension into VUI-specific fairness rules. Second, using a systematic literature review, we identified the VUI fairness design characteristics to operationalise the VUI fairness rules. Our systematic multimethod approach yields an integrative framework for fair VUIs that contextualises established fairness dimensions for VUI design, reconciles usability and fairness in voice interaction design, and explicitly equips practitioners with design characteristics to influence VA fairness. • Proposes a framework for fair VUI design that integrates fairness and usability. • Takes a holistic view using four established fairness dimensions. • Derives VUI fairness rules for each fairness dimension from expert interviews. • Derives design characteristics from literature to implement VUI fairness rules. • Supports fairness-aware VUIs through actionable and multidimensional criteria.
Institution: University of Bern
Language: English
Subfield: Human-Computer Interaction
Field: Computer Science
Domain: Physical Sciences
SDG: Reduced inequalities
Open Access status: gold
Estimated APC paid: $1,880
OpenAlex ID: W7117365794