Jing-Tong (Roger) Tzeng

I am a Master’s student at National Tsing Hua University (College of Semiconductor Research), advised by Prof. Chi-Chun (Jeremy) Lee. During my master’s studies, I also visited Prof. Carlos Busso’s Multimodal Speech Processing (MSP) Lab at The University of Texas at Dallas.

My work focuses on emotion recognition and biosignal analysis, with an emphasis on reducing aliasing introduced by front-end deep-learning preprocessing to improve downstream performance. In parallel, I develop human-in-the-loop generative components that enable targeted intervention in model outputs, increasing transparency and user trust. Broadly, I’m interested in multi-task learning and multimodal interaction.