Generative AI in Game Sound Design: Creative Practice and Collaboration
Junyi’s PhD research explores how generative AI shapes sound design as a creative and collaborative practice. As generative technologies introduce new ways of producing sonic material, they also raise questions about how creative intentions are expressed, how sounds are interpreted and refined, and how creative control is negotiated. The research approaches sound design as a situated process in which ideas develop through listening, experimentation, and interactions between people, tools, and the environments in which sound is experienced.
Situated within human-computer interaction and design research, the project investigates the relationship between AI-generated material and the practices through which it becomes meaningful and usable. It considers how practitioners incorporate generative AI into established workflows, how people with different forms of expertise communicate and develop sound ideas together, and how creative tools can support exploration while preserving agency. Game sound production provides a central context for this inquiry, bringing together sound’s expressive qualities, its responsiveness to interaction, and the collaborative demands of production.
Methodologically, the research combines studies of professional practice, prototype development, and collaborative workshops. These activities inform an iterative process of understanding existing practices and exploring possibilities for future tools. The work pays particular attention to how creative intentions evolve as sounds are generated, shared, edited, and experienced in context, and how these changes shape what people need from AI.
The research contributes empirical insights and design guidance for creativity support tools that connect generative capabilities with the wider processes of creative work. Findings on professional workflows and practitioners’ requirements for AI integration have been published at ACM Creativity & Cognition 2026. Ongoing work extends this inquiry into collaborative sound design, examining how AI can support shared exploration and the development of sonic ideas into interactive experiences.
Key details
School, Centre or Area
Area of expertise
Supervisors
Personal links
More about Junyi
Biography
Junyi Zeng is a designer and researcher working at the intersection of human-computer interaction, generative AI, and sound design. Junyi is interested in how people make sense of emerging technologies and incorporate them into creative practices, bringing a perspective grounded in design practice and attention to people’s experiences.
Degrees
2023 - MS Human-Computer Interaction, Georgia Institute of Technology
2021 - BA (Hons) Product Design, Central Saint Martins, University of the Arts London
Experience
2022 - UX Design Intern, ASML, San Jose, CA, USA
2021 - Student Researcher, Georgia Tech Sonification Lab, Atlanta, GA, USA
2020 - UX Design Intern, BMW Group, Shanghai, China,
Publications
Zeng, J., Wang, T., Diels, C. and Asadipour, A., 2026, July. Generative AI in Game Sound Design: Practitioner Workflows, Challenges, and a Design Framework. In Proceedings of the 2026 Conference on Creativity and Cognition (pp. 1337-1342).
Xu, M., Zhao, Z., Zeng, J., Wang, T., Panetta, F., Fiebrink, R., Bryan-Kinns, N. and Grierson, M., 2026, April. Glanceable Verification Indicators for AI-Assisted Storytelling. In Proceedings of the Extended Abstracts of the 2026 CHI Conference on Human Factors in Computing Systems (pp. 1-7).
Wang, Z., Zeng, J., Hall, A. and Gera, K., 2025. Beyond obedient tools: A tensional-field relational ontology for human-AI co-creation. International Association of Societies of Design Research 2025, pp.1-12.