NOTA: Multimodal Music Notation Understanding for Visual Large Language Model

TL;DR

This paper introduces NOTA, a large-scale multimodal music notation dataset and a visual language model, NotaGPT, which significantly advances music notation understanding across visual and textual modalities.

Contribution

The paper presents the first comprehensive multimodal music notation dataset and a specialized large language model trained on it, enabling improved music notation understanding.

Findings

NotaGPT-7B outperforms existing models in music understanding tasks.

The NOTA dataset covers diverse regions, enhancing model generalization.

The training pipeline effectively aligns visual and textual music representations.

Abstract

Symbolic music is represented in two distinct forms: two-dimensional, visually intuitive score images, and one-dimensional, standardized text annotation sequences. While large language models have shown extraordinary potential in music, current research has primarily focused on unimodal symbol sequence text. Existing general-domain visual language models still lack the ability of music notation understanding. Recognizing this gap, we propose NOTA, the first large-scale comprehensive multimodal music notation dataset. It consists of 1,019,237 records, from 3 regions of the world, and contains 3 tasks. Based on the dataset, we trained NotaGPT, a music notation visual large language model. Specifically, we involve a pre-alignment training phase for cross-modal alignment between the musical notes depicted in music score images and their textual representation in ABC notation. Subsequent training phases focus on foundational music information extraction, followed by training on music notation analysis. Experimental results demonstrate that our NotaGPT-7B achieves significant improvement on music understanding, showcasing the effectiveness of NOTA and the training pipeline. Our datasets are open-sourced at https://huggingface.co/datasets/MYTH-Lab/NOTA-dataset.