Abstract

Wireless earbuds have recently emerged as a new platform for wearable-based human sensing. Positioned advantageously at the human ear and equipped with unique sensors such as in-ear microphones, earbuds offer the capability to monitor a variety of human-related parameters. In this talk, I will share our recent progress in using in-ear microphones for human sensing applications, including gesture/activity recognition, user authentication, physiological monitoring, and speech enhancement. These sensing applications typically gain performance improvement with deep learning. However, earbuds are known with limited onboard resources in terms of memory, computation, and energy. Therefore, I will also discuss our recent work on enabling accurate and affordable deep learning on earbud-scale embedded devices.

Bio: Dr. Dong Ma is currently an Assistant Professor at Singapore Management University. He earned his PhD from the University of New South Wales, Australia, in 2019, followed by a postdoctoral fellowship at the University of Cambridge until 2021. His research interests focus on wearable computing systems, including human-machine interaction, physiological monitoring, and embedded machine learning. He has served as the program committee member in top venues in mobile systems including MobiCom, MobiSys, PerCom, and associate editor for IMWUT .