Skip to main navigation Skip to search Skip to main content

Co-creating Vocal Imaginaries with Conversational AIs

Research output: Conference Article in Proceeding or Book/Report chapterArticle in proceedingsResearchpeer-review

Abstract

The mainstream conversational AI and voice-user-interface (VUI) design paradigms reinforce a human-centered vocal space and gendered affective relations. This paper explores novel vocal imaginaries towards critical engagements with these technologies for desirable planetary futures. We used AI as a co-creation partner in generating VUI futures, conducted a workshop with seven participants to incorporate human perspectives in the co-creation, and sonified the resulting imaginaries. We then reflexively analyzed these imaginaries in consideration of interaction design paradigms. We propose four design paradigms, accompanied by open questions to problematize current VUI design practices. Our findings reveal that generative AI may reproduce normative assumptions or remain narrow in vocal qualities due to the limited diversity in its Western-centric training datasets. However, when approached as a creative and reflective thinking partner rather than a solution-oriented collaborator, generative AI holds potential for critical engagements with existing design paradigms towards novel vocal imaginaries.
Original languageEnglish
Title of host publicationProceedings of the 37th Australian Conference on Human-Computer Interaction
Number of pages18
Place of PublicationNew York
PublisherAssociation for Computing Machinery
Publication date28 Nov 2025
Pages1-18
ISBN (Print)979-8-4007-2016-1
DOIs
Publication statusPublished - 28 Nov 2025
EventAustralian Conference on Human-Computer Interaction - Sydney, Australia
Duration: 29 Nov 20253 Dec 2025
Conference number: 37

Conference

ConferenceAustralian Conference on Human-Computer Interaction
Number37
Country/TerritoryAustralia
CitySydney
Period29/11/202503/12/2025

Keywords

  • Voice-user-interface design
  • Generative AI
  • Participatory design
  • Critical design
  • Data bias in AI training datasets

Fingerprint

Dive into the research topics of 'Co-creating Vocal Imaginaries with Conversational AIs'. Together they form a unique fingerprint.

Cite this