Description In this presentation, I will present some of the recent progresses in building conversational agents, with a focus on the speech modality. I will introduce desirable properties of such systems and explain some of the key concepts, core ideas, and main technologies developed in practical systems. Speaker Bio Dr. Dong Yu is a Fellow of ACM, IEEE, and ISCA. His research spans speech processing and recognition, multi-modal interactive and agentic systems, and natural language processing  (including large language models) using deep learning techniques. He was a distinguished scientist and vice general manager at Tencent AI Lab till very recently. Before joining Tencent in 2017, he was a principal researcher at Microsoft Research in Redmond. He has received numerous prestigious awards, notably the 2025 IEEE SPS Claude Shannon-Harry Nyquist Technical Achievement Award, which acknowledges his pioneering contributions to speech recognition and deep learning. Dr. Dong Yu has served on multiple journal editorial boards, conference committees, and chaired the IEEE Speech and Language Processing Technical Committee from 2021 to 2022. Registration Link https://events.vtools.ieee.org/m/545402