Career Profile

Hello, I’m a Ph.D student at Music and Audio Computing Lab, KAIST, advised by Prof. Juhan Nam. My current research focuses on speech synthesis, audio tokenization, and music generation. I’m currently interested in efficient processing of audio by disentangling its features. I’m also interested in generative models such as GAN, discrete/continuous domain diffusion models, LLMs etc. I like writing songs and lyrics with my acoustic guitar. Some of my favorite artists include Radiohead, Yamashita Tatsuro, Velvet Underground, Byeong-woo Lee etc.

Education

Korean Advanced Institute of Science and Technology

2023 - present
Ph.D Student in Graduate School of Culture Technology

Music and Audio Computing Lab (Advisor - Juhan Nam)

Korean Advanced Institute of Science and Technology

2021 - 2023
MSc in Graduate School of Culture Technology

Music and Audio Computing Lab (Advisor - Juhan Nam)

Korean Advanced Institute of Science and Technology

2015 - 2021
BSc in Mechanical Engineering

Gyeonggi Science High School for the Gifted

2012 - 2015
Diploma, High School

Experiences

National Institute of Advanced Industrial Science and Technology

2026.05 - 2026.08
Technical Trainee

Neutune Co. Ltd.

2021.06 - 2021.09
Research Intern

Publications

SDP-Codec: A Speaker-Decoupled Speech Codec with Pitch Injection for Low-Bitrate Coding and Zero-Shot Voice Conversion [code] [demo webpage] [demo video]
Hounsu Kim, Juhan Nam
The 27th Annual Conference of the International Speech Communication Association (Interspeech), 2026 (accepted)
Neural speech codec for disentangled speech representation learning, low-bitrate coding, and zero-shot voice conversion.
Eunjin Choi, Hounsu Kim, Hayeon Bang, Taegyun Kwon, Juhan Nam
Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2026
A lightweight discrete diffusion model for lead-sheet-conditioned symbolic music generation.
Hounsu Kim, Taegyun Kwon, Juhan Nam
Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2025
Discrete diffusion-based refinement model for structured piano transcription.
Gyubin Lee, Hounsu Kim, Junwon Lee, Juhan Nam
Proceedings of the 27th International Conference on Digital Audio Effects (DAFx24), 2024
Best Student Paper Award
Controllable neural audio effect modeling.
Hounsu Kim, Soonbeom Choi, Juhan Nam
Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024
Controllable acoustic instrument synthesis using instrument-specific representations and diffusion outpainting.
Y. J. Cho, Y. H. Kim, Y. H. Jo, M. J. Kim, H. S. Kim, S. W. Kim, J. H. Park, and W. H. Lee*,
J. Kor. Powd. Met. Inst., 20.5 (2018) 376-381

Projects

ToplAIner - Acoustic Guitar Conditioned Singing Voice Generation [Poster] - 2023 Fall KAIST GCT731 Topics in Music Technology Final Project

Academic Advising

Gyubin Lee

Sep 2023 - Mar 2024
Department of Electronic Engineering

Donghoon Kang

Dec 2024 - Apr 2025
Department of Computer Science

Teaching Activities

Introduction to Culture Technology

Spring 2022
KAIST | Teaching Assistant

Musical Applications of Machine Learning

Fall 2022
KAIST | Teaching Assistant

Fundamentals of Computer Music

Fall 2023
KAIST | Teaching Assistant

Musical Applications of Machine Learning

Spring 2024
KAIST | Teaching Assistant

Fundamentals of Computer Music

Fall 2024
KAIST | Teaching Assistant

Musical Applications of Machine Learning

Spring 2025
KAIST | Teaching Assistant

Skills

Python & PyTorch· Deep Learning & Generative Models· Audio, Speech & Music Signal Processing· Representation Learning & Audio Tokenization· Diffusion Models, Flow-based Models & Neural Codecs· Experiment Management, Model Training & Evaluation· Linux Server Administration, Docker, SLURM & Git