Scholar
Sungho Lee
Google Scholar ID: 8yMXL5AAAAAJ
Ph.D. Student, Seoul National University
Audio signal processing
Follow
Homepage
↗
Google Scholar
↗
Citations & Impact
All-time
Citations
160
H-index
6
i10-index
4
Publications
15
Co-authors
19
list available
Contact
GitHub
Open ↗
LinkedIn
Open ↗
Publications
5 items
Rethinking Speech Representation Aggregation in Speech Enhancement: A Phonetic Mutual Information Perspective
2026
Cited
0
Reverse Engineering of Music Mixing Graphs with Differentiable Processors and Iterative Pruning
2025
Cited
0
Differentiable Acoustic Radiance Transfer
2025
Cited
0
Few-step Adversarial Schr""{o}dinger Bridge for Generative Speech Enhancement
2025
Cited
0
TokenSynth: A Token-based Neural Synthesizer for Instrument Cloning and Text-to-Instrument
2025
Cited
0
Resume (English only)
Academic Achievements
Published numerous papers in journals and conferences including JAES (2025), IEEE/ACM TASLP (2022), Interspeech (2025), ICASSP (2025, 2023), DAFx (2024), EMNLP Findings (2024), WASPAA (2023), ACM MM (2023), ISMIR LBD (2024, 2023), etc.
Recipient of Best Show & Tell Award at DAFx 2024
Best Student Paper Award Nominee at Interspeech 2025
Co-developed GRAFX, an open-source PyTorch library for audio processing graphs
Holds a Korean domestic patent: 'Multi-modal Guitar Performance Transcription Apparatus and Method' (2025)
Multiple preprints on arXiv covering differentiable acoustic radiance transfer, speech representation aggregation, and blind non-linear forward/inverse problems
Research Experience
Research Intern at Sony AI (Summer 2023): Worked on differentiable signal processing, music mixing, and graph learning
Research Intern at Supertone (Spring 2023): Focused on impulse response estimation for audio post-production; contributed to product Supertone Air
Research Project Lead at Supertone (2024): Led studio-quality speech enhancement project
Research Project Lead at Stra (2022): Developed real-time audiovisual guitar tablature system
Applicant for MCST (Korean Government) funded project (2022–2023): Musical signal processing
Applicant for IITP (Korean Government) funded project (2021–2023): SED/DOA and room acoustics analysis
Background
Ph.D. Candidate at the Music and Audio Research Group, Seoul National University
Research interests focus on inverse problems for audio
Domain expertise in room acoustics and audio effects
Methodology combines differentiable signal processing and generative models
Applications target audio post-production
Co-authors
19 total
Kyogu Lee
Professor, Seoul National University
Co-author 2
Marco A. Martinez-Ramirez
Music technology researcher, Sony AI
Stefan Uhlich
Sony Europe B.V., ZNL Deutschland
Yuki Mitsufuji
Distinguished Engineer, Sony
WeiHsiang Liao
Sony Research Inc.
Co-author 7
Co-author 8
×
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