Jupyter notebook for the automatic chord estimation metric described in:
Devaney, J. 2021. Beyond chord vocabularies: Exploiting pitch-relationships in a chord estimation metric. Late Breaking Demo. International Society for Music Information Retrieval Conference.
Abstract: Chord estimation metrics treat chord labels as independent of one another. This fails to represent the pitch relationships between the chords in a meaningful way, resulting in evaluations that must make compromises with complex chord vocabularies and that often require time-consuming qualitative analyses to determine details about how a chord estimation algorithm performs. This paper presents an accuracy metric for chord estimation that compares the pitch content of the estimation chords against the ground truth that captures both the correct notes that are estimated and additional notes that are inserted into the estimate. This is not a stand-alone evaluation protocol but rather a metric that can be integrated into existing evaluation approaches.
Paper and conference poster also included in the repository.