Jingyue Huang

Ph.D. student in Computer Science, UC San Diego

my_pic.JPG

Hi! I am the second-year Ph.D. student in Computer Science at the UC San Diego, fortunate to be advised by Prof. Julian McAuley. My research interests focus on Controllable Music Generation in both symbolic and signal domains, Music Representation Learning and Singing Voice Synthesis.

During the summer of 2025, I had the pleasure of interning at Smule, where I explored the modeling of singing performance styles. Earlier, from 2023 to 2024, I worked as a research assistant under the wonderful guidance of Prof. Yi-Hsuan Yang at National Taiwan University, focusing on emotion-driven music generation.

Before joining UCSD, I earned my M.S. degree in Data Science from New York University (NYU) and my B.S. degree in Data Science from Fudan University. During my time at Fudan, I worked with Prof. Deqing Yang on concept extraction and generation for knowledge graphs. Additionally, I worked as a research assistant at National University of Singapore with Prof. Yang You, focusing on multi-modal question answering.

Selected publications

  1. ICASSP (submitted)
    StylePitcher: Generating Style-Following and Expressive Pitch Curves for Versatile Singing Tasks
    Jingyue Huang, Qihui Yang, Fei-Yueh Chen, Julian McAuley, Randal Leistikow, Perry R. Cook, and Yongyi Zang
    International Conference on Acoustics, Speech, and Signal Processing, 2026
    PDF Website
  2. ICASSP (submitted)
    MuseTok: Symbolic Music Tokenization for Generation and Semantic Understanding
    Jingyue Huang, Zachary Novack, Phillip Long, Yupeng Hou, Ke Chen, Taylor Berg-Kirkpatrick, and Julian McAuley
    International Conference on Acoustics, Speech, and Signal Processing, 2026
    PDF Code Website
  3. ISMIR
    Emotion-Driven Piano Music Generation via Two-stage Disentanglement and Functional Representation
    Jingyue Huang, Ke Chen, and Yi-Hsuan Yang
    25th International Society for Music Information Retrieval Conference, 2024
    arXiv Code Website