Kang-wook Kim

I am a research intern at the Vision & Learning Lab, having earned my Bachelor's degree from Seoul National University. My research focuses on spoken language models and spoken dialogue systems.

Email  /  CV  /  Google Scholar  /  Semantic Scholar  /  Twitter  /  Github  /  Linkedin

* indicates equally contributed. Some of my notable papers are highlighted.

We present SubAlign, the first speech tokenization framework to explicitly segment speech at the subword level corresponding to LLM vocabularies.

We introduce an explicit "verbalization" step that translates model thoughts into speech-friendly utterances for spoken dialogue systems, along with ReVerT, an efficient verbalization model.

We present Behavior-SD and BeDLM, enabling large language models to generate natural, full-duplex spoken dialogues enriched with human conversational behaviors.

I enhanced χ-sepnet with physics-informed unrolling to improve MRI susceptibility mapping accuracy but withheld ISMRM 2024 submission due to the model's underestimation of susceptibility in patient data, requiring further refinement.

2nd place on the NTIRE Learning Super-Resolution Space Challenge 4X track and 1st place on the 8X track.

Our team developed a multilingual system that generates lip-synced talking face videos from text in four languages while preserving speaker identity.

We propose Assem-VC, a voice conversion system that combines modern techniques for realistic any-to-many conversion while preserving rhythm and intonation.

We propose a controllable singing decomposition system that encodes time-aligned linguistic content, pitch, and source speaker identity via Assem-VC.