Liang-Yuan 'Leo' WuView profile
Researcher
Liang-Yuan 'Leo' Wu is a Researcher at the University of Michigan's Computer Science and Engineering department, working with Prof. Dhruv 'DJ' Jain in the Soundability Lab at the AI Laboratory. He recently completed his Master's degree in Computer Science & Engineering at the University of Michigan. His educational background includes: Master of Science in Computer Science & Engineering, University of Michigan (2022-Present) University of Edinburgh (2021) Bachelor's degree, National Taiwan University (2017-2021) Wu's research centers on human-centered AI solutions for auditory accessibility, with deep collaboration with the Deaf and Hard of Hearing (DHH) community. He develops technologies that leverage multimodal AI and large language models to interpret soundscapes, generate personalized audio descriptions, and enhance captioning systems—particularly in challenging environments like clinical settings where communication accuracy is critical. His work bridges technical innovation with real-world user needs through mixed-methods UX research. His publication trajectory reveals a strategic focus on applying cutting-edge AI models to solve accessibility gaps in sound interpretation and captioning, with increasing emphasis on healthcare applications and community-driven design principles. This represents a significant shift toward context-aware, deployable accessibility tools rather than theoretical frameworks. Wu's research impact is recognized through: BEST POSTER AWARD at ASSETS 2024 for CARTGPT Google Academic Research Award for 'Audio Scene Understanding' proposal While not yet mentoring formal advisees, Wu secures competitive research funding through awards like Google's Academic Research Award and actively collaborates with interdisciplinary teams across HCI, AI, and accessibility domains. His work in the Soundability Lab emphasizes community co-creation with DHH individuals to ensure technologies address authentic user needs rather than theoretical scenarios. The Soundability Lab serves as Wu's primary research environment, focusing on making sound universally accessible through AI-driven innovation. The lab maintains direct partnerships with the DHH community throughout the research lifecycle—from problem identification to solution validation—ensuring technologies are both technically robust and socially impactful.












