Advancing AI for Disordered Speech: Assessment, Accessibility, and Articulation
Abstract:
Speech is the most natural interface we have, yet millions of speakers with disordered speech are locked out of it. In this talk, we discuss how AI can unlock that door and transform the landscape via "3A-dimensions" of research (Assessment, Accessibility, and Articulation). First, we address Assessment, introducing reliable algorithms that provide clinicians with objective severity diagnoses. These methods offer data-driven insights to better track patient status. Second, we focus on Accessibility by developing privacy-sensitive learning algorithms to realize automated speech recognition (ASR) systems tailored to disordered speakers. This allows users to navigate a world increasingly dominated by voice-activated services and smart environments. Finally, we explore Articulation through Voice Repair technology, which transforms unintelligible or impaired speech into clear, natural-sounding audio while preserving the speaker’s unique identity. By integrating clinical precision with functional communicative speech AI technology, this research aims to restore agency to disordered speakers, ensuring they are not only assessed and heard but understood in the modern digital landscape.
Bio:
Chi-Chun Lee (Jeremy) is a Professor and Associate Chair of the Department of Electrical Engineering at National Tsing Hua University (NTHU), Taiwan. He also serves as the Director of the NVIDIA–NTHU Joint Innovation Center and leads the Behavioral Informatics & Interaction Computing (BIIC) Lab. His research focuses on speech and language processing, affective computing, health analytics, and multimodal learning. An IEEE Senior Member and active contributor to the speech community, he serves as Associate Editor-in-Chief for IEEE Transactions on Affective Computing and Associate Editor for IEEE/ACM Transactions on Audio, Speech, and Language Processing. His leadership roles include serving as General Chair for IEEE ASRU 2023 and Area Chair for several INTERSPEECH conferences. His contributions have been recognized with numerous prestigious honors, including the 2024-2025 NTHU-Novatek Distinguished Talent Chair, NSTC Outstanding Research Award, CIEE Outstanding Electrical Engineering Professor award, IICM K. T. Li Cornerstone Award, and the Young Innovator Award from the Foundation of Outstanding Scholars. His team has won multiple INTERSPEECH challenges and best paper awards at INTERSPEECH, IEEE EMBC, and APSIPA ASC. Beyond academia, he is the co-founder of the startup AHEAD Medicine and has served as a technical consultant for major financial and technology institutions. His research has been featured in Discovery and Scientific American.