Welcome to
Human Interface Laboratory
Established in 1998 and directed by Prof. N.S.Kim, the Human Interface Laboratory conducts research on speech and audio signal processing. Current ongoing research topics include speech recognition, speech synthesis, speech enhancement, realistic acoustics, acoustic event detection, audio source seperation, and audio source localization with applications from machine learning. Not only focusing on speech and audio, this lab also covers the research in the field of Large Language Models.
About Us

서울대학교 휴먼인터페이스 연구실은 1998년 설립되어 2007년 국가 지정 연구실로 선정되었습니다. 음성 신호 처리 및 그 응용에 대한 연구를 수행하고 있으며 대표적으로 음성 인식, 합성, 화자 인식, 향상 등의 분야에서 활발하게 연구하고 있습니다.

최근 논문
Cons-KD: Dropout-Robust Knowledge Distillation for CTC-Based Automatic Speech Recognition

J. W. Yoon, H. Lee, J. Y. Kang, and N. S. Kim

IEEE Access, vol. 12, pp. 131136-131146

2024

Efficient Parallel Audio Generation Using Group Masked Language Modeling

M. Jeong, M. Kim, J. Y. Lee, and N. S. Kim

IEEE Signal Processing Letters, vol. 31, pp 979-983

2024

HuBERT-EE: Early Exiting HuBERT for Efficient Speech Recognition

J. W. Yoon, B. J. Woo, and N. S. Kim

Proc. Interspeech

Sep. 2024

MakeSinger: A Semi-Supervised Training Method for Data-Efficient Singing Voice Synthesis via Classifier-free Diffusion Guidance

S. Kim, M. Jeong, H. Lee, M. Kim, B. J. Choi, and N. S. Kim

Proc. Interspeech

Sep. 2024