李祈均 Chi-Chun Lee

開發機器學習與訊號分析演算法 推進前瞻人本應用

Imagine humans as complex systems: systems that are characterized by multiple interacting layers of hidden states (e.g., involving functions of cognition, perception, production, affect, and social interaction) generating multimodal measurable signals (speech, physiology, gestures, facial expressions, digital footprints, etc.). This paradigm shifts of taking humans within a signals and systems framework has driven research effort in developing computational methods for deriving human-centered analytics and position cross-cutting interdisciplinary research fields, e.g., affective computing, social signal processing, and behavioral signal processing. My research falls under this grant vision and centers on developing signal processing and machine learning methods of multimodal machine intelligence for human-centered analytics.

I focus on technical areas of speech processing, affective computing, health analytics, and behavior signal processing. I have contributed broadly to topics such as speech emotion modeling, spoken dialogs understanding, automated deception detection, speech analysis of autism, and ML-based clinical data analytics. I am passionate about this human-centered computational paradigm and strive continuously to advance signal processing/machine learning methods not only to enable AI systems of the modern life but also to contribute to the scientific insights of human behaviors, affective phenomena, and health and well-beings.