Education
Sep 2010 - Feb 2019
Combined M.S. and Ph.D., Electrical and Electronic Engineering, Yonsei University, Seoul, Korea
• Dissertation: Improved time-frequency trajectory excitation vocoder for deep learning-based statistical parametric speech synthesis system
• Advisor : Prof. Hong-Goo Kang
Mar 2006 - Aug 2010
B.S., Electrical and Electronic Engineering, Yonsei University, Souel, Korea
Work Experience
Jan 2023 - Present
Senior research scientist, Voice Synthesis team lead, Naver Cloud, Seongnam, Korea
• Research and development of deep learning-based TTS models
• Implementing TTS api for cloud services such as Clova Voice Pro, Clova Dubbing, and Voice Maker
• Specialities: High-quality neural vocoders and controllable TTS models
Aug 2022 - Present
Adjunct professor, Artificial Intelligence Institute, SNU, Seoul, Korea
Mar 2017 - Dec 2022
Senior research scientist, Voice Model team lead, Clova Voice, Naver Corp., Seongnam, Korea
• Research and development of TTS system combining deep learning and unit-selection TTS models
• Implementing cloud-based real-time TTS products for Clova AI speaker, Maps navigation, and News anchor
Aug 2016 - Nov 2016
Research Intern, Qualcomm Technologies Inc., San Diego, CA
• Spatial audio: Fixed-point implementation of MPEG-H 3D Audio Decoder
• Mentor: Dr. Deep Sen
Sep 2015 - Feb 2016, Apr 2016 - Jun 2016
Research Intern, Microsoft Research Asia, Beijing, China
• Speech synthesis: Deep learning-based TTS system using ITFTE vocoder
• Mentor: Dr. Frank Soong
Academic Activites
Reviewer
• Signal Processing Letters 2023
• INTERSPEECH 2020 - 2024
• ICASSP 2021 - 2024
Honors and Awards
• Innovators Under 35 Korea, MIT Technology Review, Dec 2022
• Ranked No. 2, N Innovation Award 2020, Naver Corp., Dec 2020
• The Best Paper Award, APSIPA ASC 2020, Dec 2020
• Ranked No. 1, N Innovation Award 2019, Naver Corp., Dec 2019
• Ranked No. 1, N Innovation Award 2018, Naver Corp., Nov 2018
• Excellent Intern Award, Microsoft Research Asia, Jun 2016
• Excellent Intern Award, Microsoft Research Asia, Feb 2016
Patents
• KR10-2661751, Method and system for generating speech synthesis model based on selective data augmentation, Apr 2024
• KR10-2626618, Method and system for synthesizing emotional speech based on emotion prediction, Jan 2024
• KR10-2621842, Method and system for non-autoregressive speech synthesis, Jan 2024
• KR10-2198598, Method for generating synthesized speech signal, neural vocoder, and training method thereof, Dec 2020
• KR10-2198597, Neural vocoder and training method of neural vocoder for constructing speaker-adaptive model, Dec 2020