From Discord to Harmony: Decomposed Consonance-based Training for Improved Audio Chord Estimation
This work addresses annotator subjectivity and class imbalance in ACE, which is important for music transcription and analysis, but it is incremental as it builds on existing conformer models with specific enhancements.
The paper tackles challenges in Audio Chord Estimation (ACE) by evaluating inter-annotator agreement with new metrics and proposing a consonance-informed distance metric to better capture perceptual similarity, resulting in a novel conformer-based model that integrates consonance-based label smoothing and decomposes chord estimation to address class imbalance.
Audio Chord Estimation (ACE) holds a pivotal role in music information research, having garnered attention for over two decades due to its relevance for music transcription and analysis. Despite notable advancements, challenges persist in the task, particularly concerning unique characteristics of harmonic content, which have resulted in existing systems' performances reaching a glass ceiling. These challenges include annotator subjectivity, where varying interpretations among annotators lead to inconsistencies, and class imbalance within chord datasets, where certain chord classes are over-represented compared to others, posing difficulties in model training and evaluation. As a first contribution, this paper presents an evaluation of inter-annotator agreement in chord annotations, using metrics that extend beyond traditional binary measures. In addition, we propose a consonance-informed distance metric that reflects the perceptual similarity between harmonic annotations. Our analysis suggests that consonance-based distance metrics more effectively capture musically meaningful agreement between annotations. Expanding on these findings, we introduce a novel ACE conformer-based model that integrates consonance concepts into the model through consonance-based label smoothing. The proposed model also addresses class imbalance by separately estimating root, bass, and all note activations, enabling the reconstruction of chord labels from decomposed outputs.