Najim Dehak, associate professor of electrical and computer engineering at Johns Hopkins Whiting School of Engineering, along with co-authors Zheng-Hua Tan and Achintya Kr. Sarkar of Aalborg University, Denmark, received the Best Research Paper award at the International Symposium on Computer Architecture (ICSA) conference in Dublin, Ireland in August. Their paper presents an unsupervised segment-based method for robust voice activity detection (rVAD), offering the potential to advance voice processing and enhance user experiences. This technology has the potential to improve public safety by helping identify emergencies and facilitating emergency responses in noisy public settings, and in healthcare through improved functionality of voice-controlled medical devices and telemedicine in noisy hospital environments.
Dehak’s work has impacted speech processing, audio classification, and health applications, gaining significant recognition in the speech community.