MID-FiLD: MIDI Dataset for Fine-Level Dynamics


  • Jesung Ryu Pozalabs, Republic of Korea
  • Seungyeon Rhyu Pozalabs, Republic of Korea
  • Hong-Gyu Yoon Pozalabs, Republic of Korea
  • Eunchong Kim Pozalabs, Republic of Korea
  • Ju Young Yang Duke University, United States
  • Taehyun Kim Pozalabs, Republic of Korea




APP: Other Applications, ML: Applications


One of the challenges in generating human-like music is articulating musical expressions such as dynamics, phrasing, and timbre, which are difficult for computational models to mimic. Previous efforts to tackle this problem have been insufficient due to a fundamental lack of data containing information about musical expressions. In this paper, we introduce MID-FiLD, a MIDI dataset for learning fine-level dynamics control. Notable properties of MID-FiLD are as follows: (1) All 4,422 MIDI samples are constructed by professional music writers with a strong understanding of composition and musical expression. (2) Each MIDI sample contains four different musical metadata and control change \#1 (CC\#1) value. We verify that our metadata is a key factor in MID-FiLD, exerting a substantial influence over produced CC\#1 values. In addition, we demonstrate the applicability of MID-FiLD to deep learning models by suggesting a token-based encoding methodology and reveal the potential for generating controllable, human-like musical expressions.



How to Cite

Ryu, J., Rhyu, S., Yoon, H.-G., Kim, E., Yang, J. Y., & Kim, T. (2024). MID-FiLD: MIDI Dataset for Fine-Level Dynamics. Proceedings of the AAAI Conference on Artificial Intelligence, 38(1), 222-230. https://doi.org/10.1609/aaai.v38i1.27774



AAAI Technical Track on Application Domains