Chao Huang

Building AI that can listen, speak, reason, and interact in real time.

Research Scientist, Tencent Hunyuan · Seattle

LLM-based ASR full-duplex models interactive multimodal models

At Tencent Hunyuan, I work on LLM-based speech recognition, full-duplex spoken interaction, and multimodal interaction models. My research focuses on building AI systems that can understand, respond to, and interact with people naturally in real time.

More broadly, my work connects omni-modal LLMs, multimodal reasoning, audio-visual learning, and diffusion-based generation, with an emphasis on learning more from less data and computation.

Ph.D. in Computer Science, University of Rochester · B.Eng., Nanjing University