International Journal of Secondary Computing and Applications Research


home | blog | events | pubs | scholarship

Leveraging LLMs for Automated MIDI Generation

Corey Zhang, Mark Santolucito

Affiliation: Eastlake High School

IJSCAR Vol. 1, Issue 1 (2024)  ·  pp. 3–10

DOI: 10.5281/zenodo.14279480


Abstract

This paper explores the application of large language models (LLMs) for music generation specifically focusing on generating MIDI files JSON representations and music-related code. We aim to leverage LLM capabilities to automate and enhance the music composition process. We systematically examine methodologies and tools for integrating LLMs in music generation with direct music creation and code-based techniques. The challenge of automating music composition using AI remains significant due to the complexity and creativity required in the process. Our findings demonstrate the potential of LLMs to innovate and streamline music composition offering new tools and approaches for musicians and developers.


Keywords: large language models, music generation, code generation


View Full Issue PDF   All Publications