In the field of automatic symbolic music analysis, significant progress has been made, yet existing systems are typically designed for a single mode of use, such as full-score prediction, which fails to accommodate the broader operational needs that arise in analysis workflows, including partial completion, local correction, and iterative refinement. This results in a gap between strong benchmark models and systems that can support interactive analytical use.
We present a unified framework for symbolic Roman-numeral (RN) analysis that narrows this gap by combining strong predictive performance with direct support for constrained completion and revision. The method aims to provide a practical trade-off between accuracy and interactive responsiveness by computing expensive pretrained representations once and reusing them during iterative refinement, making powerful pretrained models more amenable to interactive settings.
The framework supports complete score analysis, targeted revision of existing labels, and inference of missing annotations from partial context through a shared modeling framework. Experiments on Dilemmadata, the largest and most heterogeneous benchmark of its kind, demonstrate that the proposed approach serves as a strong RN-analysis baseline while also supporting masked completion from partial labels. Together with a prototype interface for multi-level candidate inspection and editing, these results position automatic RN analysis not merely as a prediction problem, but as a foundation for future interactive tools in music analysis.
Blogger's Review: This paper illustrates how to transform powerful predictive models into practical interactive tools, successfully bridging the gap between existing technology and user needs. By optimizing the iterative process, it not only enhances the accuracy of music analysis but also increases user engagement, showcasing a promising future for interactive music analysis tools.