Symbolic Music Generation with Fine-grained Interactive Textural Guidance

Tingyu Zhu, Haoyu Liu, Zhimin Jiang, Zeyu Zheng

Keywords: Symbolic Music Generation; Guided Diffusion Models

Abstract: The problem of symbolic music generation presents unique challenges due to the combination of limited data availability and the need for high precision in note pitch. To address these issues, we introduce an efficient Fine-grained Sampling Guidance (FTG) approach within diffusion models. FTG guides the diffusion models to generate music that aligns more closely with the control and intent of human composers, thereby improving the accuracy and quality of music generation. This method empowers diffusion models to excel in advanced applications such as improvisation, and interactive music creation. We derive theoretical characterizations for both the challenges in symbolic music generation and the effect of the FTG approach. We provide numerical experiments, subjective evaluation and a demo page for interactive music generation with user input to showcase the effectiveness of our approach.

Primary Area: applications to computer vision, audio, language, and other modalities

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 11131