Improving the Classification of Rare Chords with Unlabeled Data
This work addresses the problem of accurately classifying rare chords for musicians and music information retrieval systems, offering an incremental improvement to existing ACR methods.
This paper explores methods to enhance the classification of rare chords in Automatic Chord Recognition (ACR). The authors found that a self-learning technique with noise addition yielded the best results for improving rare chord recognition.
In this work, we explore techniques to improve performance for rare classes in the task of Automatic Chord Recognition (ACR). We first explored the use of the focal loss in the context of ACR, which was originally proposed to improve the classification of hard samples. In parallel, we adapted a self-learning technique originally designed for image recognition to the musical domain. Our experiments show that both approaches individually (and their combination) improve the recognition of rare chords, but using only self-learning with noise addition yields the best results.