Scholar
Kang-wook Kim
Google Scholar ID: DZ08mDcAAAAJ
Seoul National University
Speech and Language Processing
Speech Synthesis
Follow
Homepage
↗
Google Scholar
↗
Citations & Impact
All-time
Citations
106
H-index
4
i10-index
4
Publications
8
Co-authors
8
list available
Contact
Email
full324@snu.ac.kr
CV
Open ↗
Twitter
Open ↗
GitHub
Open ↗
LinkedIn
Open ↗
Publications
1 items
Think, Verbalize, then Speak: Bridging Complex Thoughts and Comprehensible Speech
2025
Cited
0
Resume (English only)
Research Experience
Conducting research on spoken dialogue systems and language models at Vision & Learning Lab
Developed ReVerT, a verbalization model that translates internal model thoughts into speech-friendly utterances
Proposed Behavior-SD and BeDLM frameworks enabling LLMs to generate natural, full-duplex spoken dialogues with human conversational behaviors
Contributed to a multilingual TTS-based talking face generation system
Proposed Assem-VC, a realistic any-to-many voice conversion system preserving rhythm and intonation
Built a controllable and interpretable singing voice decomposition system based on Assem-VC
Co-authors
8 total
Junhyeok Lee
Johns Hopkins University, Center for Language and Signal Processing
Dongseok Shim
Sony
Ki-Ung Song
AI Research Engineer at Nota AI
Hyoung-Kyu Song
Applied Scientist (Member of Technical Staff) @ Mirage
Co-author 5
Gunhee Kim
Professor, Seoul National University
Sehun Lee
Vision and Learning Lab, SNU
Sang Hoon Woo
Independent Researcher
×
Welcome back
Sign in to Agora
Welcome back! Please sign in to continue.
Email address
Password
Forgot password?
Continue
Do not have an account?
Sign up