01 / Updates

News

2026

  1. Presented MeloDISinger at the Interspeech 2026 poster session 🐨.

  2. Our system ranked first in Track 2 of the VoiceMOS Challenge 2026 for predicting speech naturalness and emotion similarity in emotional TTS! πŸ₯‡.

  3. Research stay at Max Planck Institute for Empirical Aesthetics, working on a survey of empathetic and conversational spoken language systems πŸ‡©πŸ‡ͺ.

  4. Gave a research talk on MeloDISinger at the AIR Lab Seminar, University of Rochester πŸ‡ΊπŸ‡Έ.

  5. MeloDISinger, my first M.S. work on singing voice editing, was accepted to Interspeech 2026 πŸŽ‰.

  6. Awarded a government-funded research training scholarship (USD 37K) and joined the University of Toronto as a visiting graduate student πŸ‡¨πŸ‡¦.

2025

  1. Volunteered at ISMIR 2025 and had a great time meeting the MIR community 🎢.

  2. Served as student coordinator for the KSMI 2025 Founding Symposium πŸ‡°πŸ‡·.

2024

  1. CoT-Sep, my first paper, was accepted as an oral presentation at IEEE FLLM 2024. πŸŽ‰

  2. Started my M.S. at KAIST AI. πŸŽ“

02 / Background

Education

Korea Advanced Institute of Science and Technology

Feb. 2024 β€” Feb. 2027

M.S. Student, Graduate School of Artificial Intelligence

Music and Audio Computing Lab Β· Advisor: Prof. Juhan Nam

University of Toronto

Jan. 2026 β€” Apr. 2026

Visiting Graduate Student, Mechanical & Industrial Engineering

IITP-affiliated collaborative AI research program at LG Electronics Toronto AI Lab

Yonsei University

Mar. 2019 β€” Aug. 2024

B.S. in Computer Science

Overall GPA: 3.8/4.3 Β· CS GPA: 3.96/4.3

University of British Columbia

Jan. 2023 β€” Apr. 2023

Exchange Student, Computer Engineering

Relevant course: Topics in Computer Engineering β€” Deep Learning

03 / Selected work

Publications (*: Equal contribution)

Figure from UTMOST: A Multi-Task Model for Assessing Emotional Speech Synthesis

Under review

UTMOST: A Multi-Task Model for Assessing Emotional Speech Synthesis

Joonyong Park*, Hounsu Kim*, Yoonjeong Park*, Ryota Kawamatsu, Wataru Nakata, Yuki Saito, Satoru Fukayama, Juhan Nam.

Figure from FOAdapter: A Plug-and-Play Spatial Tokenizer for Low-Bitrate FOA Speech Coding

Under review

FOAdapter: A Plug-and-Play Spatial Tokenizer for Low-Bitrate FOA Speech Coding

Kirak Kim*, Yoonjeong Park*, Juhan Nam, Sungyoung Kim, Minje Kim.

Figure from MeloDISinger: Melody-Aware & Duration-Preserving Singing Voice Editing with Audio Infilling

2026 Β· Interspeech

MeloDISinger: Melody-Aware & Duration-Preserving Singing Voice Editing with Audio Infilling

Yoonjeong Park, Jaekwon Im, Juhan Nam.

Interspeech 2026 Β· Poster

Figure from Can Separators Improve Chain-of-Thought Prompting?

2024 Β· IEEE FLLM

Can Separators Improve Chain-of-Thought Prompting?

Yoonjeong Park*, Hyunjin Kim*, Chanyeol Choi, Junseong Kim, Jy-yong Sohn

IEEE International Conference on Foundation and Large Language Models (FLLM), 2024 Β· Oral Presentation