BACHI: Boundary-Aware Symbolic Chord Recognition Through Masked Iterative Decoding on Pop and Classical Music

2025-10-07

AI Summary

Symbolic music automatic chord recognition (ACR) has long suffered from scarce annotated data and misalignment between model design and human music analysis practices. To address these challenges, we propose BACHI, a boundary-aware iterative decoding framework. First, we construct POP909-CL—a beat-aligned, manually corrected dataset derived from POP909. Second, we introduce a multi-stage decoding architecture that decomposes chord recognition into sequential inference of root, quality, and bass—mimicking human auditory training. Third, we integrate a boundary detection network with masked iterative decoding and a music-theory-informed joint loss function. BACHI is the first symbolic-domain ACR method to jointly optimize chord boundary localization and structural inference. It achieves state-of-the-art performance on both popular and classical music benchmarks. Ablation studies confirm the efficacy of each component, demonstrating significant accuracy gains—particularly in complex tonal and rhythmic contexts.

Technology Category

• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition
• Planning, Routing, and Scheduling: Activity and Plan Recognition
• Cognitive Modeling & Cognitive Systems: Symbolic Representations

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Automatic chord recognition (ACR) via deep learning models has gradually achieved promising recognition accuracy, yet two key challenges remain. First, prior work has primarily focused on audio-domain ACR, while symbolic music (e.g., score) ACR has received limited attention due to data scarcity. Second, existing methods still overlook strategies that are aligned with human music analytical practices. To address these challenges, we make two contributions: (1) we introduce POP909-CL, an enhanced version of POP909 dataset with tempo-aligned content and human-corrected labels of chords, beats, keys, and time signatures; and (2) We propose BACHI, a symbolic chord recognition model that decomposes the task into different decision steps, namely boundary detection and iterative ranking of chord root, quality, and bass (inversion). This mechanism mirrors the human ear-training practices. Experiments demonstrate that BACHI achieves state-of-the-art chord recognition performance on both classical and pop music benchmarks, with ablation studies validating the effectiveness of each module.

Problem

• Addresses symbolic chord recognition challenges in music data
• Improves chord boundary detection and iterative ranking accuracy
• Bridges gap between computational methods and human music analysis

Innovation

• Boundary-aware symbolic chord recognition model
• Decomposed task into boundary detection and ranking
• Masked iterative decoding mirrors human ear-training practices