About me

I am a second-year Ph.D. student at KAIST Multimodal AI Lab, advised by Professor Joon Son Chung. My research focuses on multimodal AI, particularly audio-visual speech processing and source separation. I investigate how cross-modal cues can be leveraged to isolate target signals and improve speech perception in complex acoustic environments, with the goal of building robust multimodal systems for real-world applications.

Education

Selected Awards

Experience

Publications

2026

LRS-VoxMM: A benchmark for in-the-wild audio-visual speech recognition
D. Kwak, J. Choi, S. Lee, and J. S. Chung
Technical Report
Paper Project Page

I4U’s Official and Streamlined Audio Systems for NIST SRE24
D. Colibro, C. Vair, Y. Tu, J. Li, Z. Huang, Y. Chen, K. A. Lee, M. Mak, J. Mishra, V. Singh, X. Xuan, M. Chhibber, O. Kurnaz, T. Kinnunen, S. Lee, C. Jung, K. Nam, J. S. Chung, and S. Wang
Odyssey 2026
Paper

Cinematic Audio Source Separation Using Visual Cues
K. Zhang^*, S. Lee^*, A. Senocak, and J. S. Chung
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Paper Project Page

Plug-and-Steer: Decoupling Separation and Selection in Audio-Visual Target Speaker Extraction
D. Kwak^*, S. Lee^*, and J. S. Chung
Interspeech
Paper Project Page

2025

Model-Guided Dual-Role Alignment for High-Fidelity Open-Domain Video-to-Audio Generation
K. Zhang^*, T. X. Pham^*, S. Lee, A. Niu, A. Senocak, and J. S. Chung
Conference on Neural Information Processing Systems (NeurIPS)
Paper Code

2024

Seeing Through the Conversation: Audio-Visual Speech Separation based on Diffusion Model
S. Lee^*, C. Jung^*, Y. Jang, J. Kim, and J. S. Chung
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Paper Project Page Code

TalkNCE: Improving Active Speaker Detection with Talk-Aware Contrastive Learning
C. Jung^*, S. Lee^*, K. Nam, K. Rho, Y. Jang, and J. S. Chung
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Paper Code

FlowAVSE: Efficient Audio Visual Speech Enhancement Models with Conditional Flow Matching
C. Jung, S. Lee, J. H. Kim, and J. S. Chung
Interspeech
Paper Code

Contact