Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

Overcoming speech barriers: Non-verbal voice cue interaction technique for enhancing smart voice assistant accessibility for individuals with dysarthria

Jaddoh, Aisha, Loizides, Fernando ORCID: https://orcid.org/0000-0003-0531-6760, Alreafai, Khadijah and Rana, Omer ORCID: https://orcid.org/0000-0003-3597-2646 2025. Overcoming speech barriers: Non-verbal voice cue interaction technique for enhancing smart voice assistant accessibility for individuals with dysarthria. ACM Transactions on Accessible Computing 10.1145/3726874

[thumbnail of Overcoming_Speech_Barriers_Non_Verbal_Voice_CueInteraction_Technique_for_Enhancing_Smart_Voice_Assistant_Accessibility_for_Individuals_with_Dysarthria.pdf]
Preview
PDF - Accepted Post-Print Version
Download (2MB) | Preview

Abstract

Inclusive user experience design that caters to a broad range of users can enhance the ease of interaction with systems. Despite progress, certain systems like smart voice assistants often remain inaccessible to some user types, particularly individuals with dysarthria whose speech may not be recognized by these devices. In this article, we present a novel interaction technique that relies on non-verbal voice cues instead of words and sentences. To evaluate the usability, user experience, and workload of this approach, we developed a system (DARIA) and conducted two studies, exploring two different design options. The first study tested the system without a customization option, while the second introduced a customization option. We used three questionnaires and conducted post-study interviews for thorough evaluation. The findings indicated user acceptance of this interaction technique in general was high. When compared with a market leader product (Alexa), DARIA showed higher performance in terms of accessibility. Users found the non-verbal system to be more usable and likable. Interestingly, the customizable design was found less usable than the pre-defined.

Item Type: Article
Date Type: Published Online
Status: In Press
Schools: Schools > Computer Science & Informatics
Publisher: Association for Computing Machinery (ACM)
ISSN: 1936-7228
Date of First Compliant Deposit: 17 April 2025
Date of Acceptance: 14 March 2025
Last Modified: 17 Apr 2025 13:30
URI: https://orca.cardiff.ac.uk/id/eprint/177691

Actions (repository staff only)

Edit Item Edit Item

Downloads

Downloads per month over past year

Loading...

View more statistics