Keon Lee
Deep Learning Researcher at KRAFTON Inc.
Seoul, South Korea
Summary
👤
Senior🎓
Top SchoolKeon Lee is a deep learning researcher based in Seoul with 8 years of experience focused on conversational AI and generative models. Currently at KRAFTON's GenAI research division, he specializes in text-to-speech as part of a broader vision to build personality-rich agents that listen, think, and speak like humans. Previously a data scientist at LOVO AI, he brings hands-on industry experience in speech and voice technologies grounded in an MEng from KAIST. Practical yet research-minded, Keon combines industrial product exposure with academic rigor in computing and industrial design, enabling him to shape both the technical and experiential aspects of voice agents. His long-term goal—to create agents with distinctive, human-like personalities—drives his focused work on TTS as the most direct vehicle for expressing individuality in AI.
8 years of coding experience
1 year of employment as a software developer
Master of Engineering - MEng, School of Computing, Master of Engineering - MEng, School of Computing at 한국과학기술원(KAIST)
Github Skills (70)
text-to-speech10
generative10
self-attention10
speech10
text-based10
duration10
speech-synthesis10
speech-to-text10
chainer10
generative-model10
machine-translation9
styler9
midi9
mechanism9
speech-processing9
Programming languages (3)
Github contributions (5)
keonlee9420/Comprehensive-Tacotron2
Jul 2021 - Feb 2022
PyTorch Implementation of Google's Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions. This implementation supports both single-, multi-speaker TTS and several techniques to enforce the robustness and efficiency of the model.
Contributions:3 releases, 5 commits, 7 pushes in 7 months
autoregressivecomprehensivedeep-learningdiagonal-guided-attentionhifi-gan
keonlee9420/PortaSpeech
Oct 2021 - Feb 2022
PyTorch Implementation of PortaSpeech: Portable and High-Quality Generative Text-to-Speech
Contributions:1 release, 14 commits, 26 pushes in 4 months
text-to-speechspeech-recognitiongenerative-textvaenormalizing-flows
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.