Ting Dang

Senior Research Associate

University of Cambridge

Dr. Ting Dang is a Senior Research Associate under Professor Cecilia Mascolo’s supervision in Department of Computer Science and Technology, University of Cambridge, and a Visiting Research Associate in the University of New South Wales (UNSW), Sydney, Australia. She received the Ph.D. degree under supervision of Dr. Vidhyasaharan Sethu and Professor Eliathamby Ambikairajah from UNSW in 2018. Her research primarily explores audio signals (e.g., speech, cough, heart sounds collected from wearables) for mental state (e.g., emotion and depression) and disease (e.g., COVID-19) detection and tracking in a remote monitoring context. She has published 17 papers in high rank journals and conferences. She has been supervising students at different levels from Bachelor thesis to PhDs across UNSW and Cambridge, and collaborated with researchers from UNSW, University of Cambridge, University of Southampton, etc. She has been served as the Programme Committee and outstanding reviewer for many top journals and conferences. 

COVID-19 disease progression prediction and forecasting (European Research Council Grant) & Speech-based continuous emotion prediction and tracking

While most commonly used COVID-19 diagnosis tools such as PCR tests are intrusive and time-consuming, digital technologies that employ machine learning techniques for automatic COVID-19 detection and monitoring allow for fast and scalable detection. This project aims to explore the potential of longitudinal audio biomarkers to aid the disease progression prediction and forecasting ahead of time using deep learning techniques. The work has been published in Nature NPJ digital medicine, Journal of Medical Internet Research, etc., and reported by the media including BBC, HORIZON, etc

Automatic speech-based emotion prediction plays a key role in many applications such as interactive humancomputer interface design and clinical diagnosis tools. However, emotion expressions or perceptions are in general heterogeneous across individuals, depending on a wide range of factors, such as cultural background. This project aims to explore this source of variations on the emotion prediction systems and develop new paradigms to represent and model emotion that accounts for the individual differences. The work has been published in IEEE Transactions on Affective Computing, Frontiers in Computer Science, etc.