Projects

Naver Cloud, Seongnam, Korea

Nov 2022 - present
Development of High-quality TTS api for cloud services
  • Role: Management, research and development of DNN TTS models
  • Related services
     · Clova Voice Pro

May 2022 - Oct 2023
Development of automatic TTS modeling with smartphone recordings
  • Role: Management, research and development of DNN TTS models
  • Related services
     · Voice Maker

Naver Corp., Seongnam, Korea

Mar 2017 - Dec 2022
Hybrid TTS combining deep learning and unit-selection models
  • Role: Management, research and development of DNN TTS models
  • Related services
     · Showhost voice for Naver Shopping, May 2021 - present
     · Representative voice for Clova Contact Center, May 2019 - present
     · Navigation voice for Naver Map, Jan 2019 - Sep 2020
     · Sangjin Oh’s (Korean newscaster) voice for Naver News, Oct 2019 - May 2020
     · In-Na Yoo’s (Korean actress) voice for Clova AI speaker, Apr 2018 - Dec 2018

Sep 2021 - May 2022
TTS-based audiobook generation
  • Role: Development of audiobook generation tool
  • Related services
     · Clova Dubbing x Inmun 360

Yonsei University, Seoul, Korea

Nov 2016 - Mar 2017
Development of Glottal DNN vocoder for speech synthesis systems
  • Naver Corp., Korea
  • Role: Research and development of Glottal DNN vocoder for DNN-based TTS systems

Mar 2015 - Feb 2016
Deep neural network-based speaker-adaptive text-to-speech
  • ReadSpeaker (Voiceware), Korea
  • Role: Research and development of DNN-based TTS acoustic models

Jun 2015 - Aug 2015
Low bit-rate speech codec
  • Line Corp., Korea
  • Role: Implementation of low bit-rate MELP codec

Mar 2014 - Feb 2015
Statistical parametric speech synthesis
  • ReadSpeaker (Voiceware), Korea
  • Role: Research and development of parametric vocoder for DNN-based TTS systems

Feb 2012 - Dec 2014
Multi-model bio interface and recognition
  • Samsung Electronics, Korea
  • Role: Research and development of speech enhancement for pathological voices

Nov 2013 - Jun 2014
Waveform generation for statistical parametric speech synthesis
  • Microsoft Research, Korea
  • Role: Research and development of DNN-based TTS models

Aug 2011 - Jan 2012
Fixed point implementation of MPEG-D USAC decoder
  • LG Electronics, Korea
  • Role: Fixed-point implementation