Chin-Hui Lee

Timeline

1. 1973–1981

   Learning the foundations

   He earned a B.S. in electrical engineering at National Taiwan University in 1973, an M.S. in engineering and applied science at Yale in 1977, and a Ph.D. in electrical engineering with a statistics minor at the University of Washington in 1981.

2. 1981–2001

   Speech systems in industry

   He began at Verbex, where he researched connected-word recognition, then worked on speech coding, synthesis, and speech recognition at Digital Sound Corporation. From 1986 to 2001, he worked at Bell Laboratories, became a Distinguished Member of Technical Staff, and directed the Dialogue Systems Research Department.

3. 2001–2002

   From Singapore to Georgia Tech

   He was a visiting professor at the National University of Singapore from August 2001 to August 2002. He joined Georgia Tech's School of Electrical and Computer Engineering in September 2002.

4. 2006

   An award for speech recognition

   The IEEE Signal Processing Society gave him its Technical Achievement Award for contributions to automatic speech recognition.

5. 2012

   A medal and a fellowship

   ISCA awarded him its Medal for contributions to automatic speech and speaker recognition, including adaptive learning, discriminative training, and utterance verification. The association also named him an ISCA Fellow in 2012.

6. 2017

   Mapping noisy speech

   At a 2017 NUS lecture, he described mapping noisy-speech log-power spectral features to clean speech with deep neural networks. He outlined an integrated deep-learning approach for signal conditioning, feature extraction, and acoustic phone modeling in an automatic speech recognition system.

7. 2018

   Neural enhancement recognized

   The IEEE Signal Processing Society listed his coauthored paper, “A Regression Approach to Speech Enhancement Based on Deep Neural Networks,” among its 2018 Best Paper Award recipients. Yong Xu, Jun Du, and Li-Rong Dai joined him as coauthors.

8. As of October 2026

   Research across two institutions

   As of October 2026, he is a professor in Georgia Tech's School of Electrical and Computer Engineering and an honorary chair professor at Asia University, which lists his Chinese name as 李錦輝. In April 2026, Georgia Tech listed him as adviser for Chun-wei Ho's PhD proposal on audio segmentation, target music extraction, and cinematic audio source separation. His research applies machine learning to speech and language processing, and his 2025 coauthored paper studies variational Bayesian learning for deep latent variables in acoustic knowledge transfer.