Dr. Yilin Pan is a Lecturer and Master Supervisor at the School of Artificial Intelligence, Dalian Maritime University, and a part-time researcher at the Center for Ageing Language and Care, Tongji University. She received her Ph.D. in Computer Science from the University of Sheffield, UK in 2022. Her research interests include pathological speech detection, multimodal emotion analysis, and intelligent voice dialogue. Since joining Dalian Maritime University in 2022, she has hosted one project funded by the Liaoning Provincial Natural Science Foundation and participated in one General Program of the National Natural Science Foundation of China. In the past five years, she has published more than ten papers in authoritative conferences and journals in the speech field such as INTERSPEECH, ICASSP, IEEE TASLP, IEEE Signal Processing Letters, and Computer Speech and Language, with over 400 Google Scholar citations. She received the European Marie Curie Full Scholarship in 2018 and was recognized as a Dalian City Young Talent in 2022.

Early cognitive decline caused by neurodegenerative diseases such as Alzheimer's disease presents specific speech and language characteristics in patients' spontaneous speech. However, current mainstream clinical screening methods rely primarily on manual scale assessments, making it difficult to carry out large-scale routine early screening. Non-invasive intelligent detection of cognitive impairment based on patients' natural conversational speech has become a cutting-edge research direction with great application value at the intersection of speech signal processing and smart healthcare.

This talk takes automated diagnosis of cognitive health in the elderly as the core task, and conducts a series of systematic studies around pathological speech feature extraction and cross-modal fusion modeling of speech and text, completely elaborating an end-to-end intelligent diagnosis technology system based on patients' natural discourse.