2018:Audio Chord Estimation Results

Introduction

This page contains the results of the 2018 edition of the MIREX automatic chord estimation tasks. This edition was the sixth one since the reorganization of the evaluation procedure in 2013. The results can therefore be directly compared to those of the last five years. Chord labels are evaluated according to five different chord vocabularies and the segmentation is also assessed. Additional information about the used measures can be found on the page of the 2013 edition.

What’s new?

• This year the algorithms have additionally been evaluated on the "RWC-Popular" and "USPOP2002Chords" dataset annotated at the Music and Audio Research Lab of NYU, whose annotations are publicly available. The RWC-Popular dataset contains 100 pop songs recorded specifically for music information retrieval research. The USPOP2002Chords set is the 195 file subset of the USPOP2002 dataset that have been annotated with chord sequences.

Software

All software used for the evaluation has been made open-source. The evaluation framework is described by Pauwels and Peeters (2013). The corresponding binaries and code repository can be found on GitHub and the used measures are available as presets. The raw algorithmic output will be provided later More help can be found in the readme.

The statistical comparison between the different submissions is explained in Burgoyne et al. (2014). The software is available at BitBucket. It uses the detailed results provided below as input.

Submissions

	Abstract	Contributors
CM1	PDF	Chris Cannam, Matthias Mauch
JLCX1, JLCX2	PDF	Junyan Jiang, Ke Chen, Wei Li, Guangyu Xia
SG1	PDF	Franz Strasser, Stefan Gaser
FK2	PDF	Florian Krebs, Filip Korzeniowski, Sebastian Böck

Results

Summary

All figures can be interpreted as percentages and range from 0 (worst) to 100 (best).

Isophonics2009

Algorithm	MirexRoot	MirexMajMin	MirexMajMinBass	MirexSevenths	MirexSeventhsBass	MeanSeg	UnderSeg	OverSeg
CM1	78.66	75.51	72.58	54.78	52.36	85.87	87.22	85.98
FK2	87.38	86.80	83.43	75.55	72.60	89.29	87.24	92.35
JLCX1	86.75	86.25	84.44	75.87	74.39	90.33	88.36	93.38
JLCX2	86.51	86.05	84.23	75.64	74.17	90.12	87.94	93.48
SG1	82.03	78.67	76.84	69.20	67.55	82.34	89.03	77.87

download these results as csv

Billboard2012

Algorithm	MirexRoot	MirexMajMin	MirexMajMinBass	MirexSevenths	MirexSeventhsBass	MeanSeg	UnderSeg	OverSeg
CM1	74.23	72.31	70.25	55.44	53.48	83.60	85.33	83.31
FK2	86.30	86.01	83.12	61.12	58.75	88.38	86.78	90.96
JLCX1	83.09	82.79	81.48	70.88	69.72	87.72	86.65	89.85
JLCX2	83.60	83.31	82.01	71.26	70.11	88.35	86.94	90.90
SG1	82.11	79.28	78.04	66.57	65.53	81.60	90.21	76.22

download these results as csv

Billboard2013

Algorithm	MirexRoot	MirexMajMin	MirexMajMinBass	MirexSevenths	MirexSeventhsBass	MeanSeg	UnderSeg	OverSeg
CM1	71.23	67.36	65.28	49.07	47.25	81.50	83.13	82.53
FK2	80.60	78.37	75.85	55.81	53.57	83.59	83.01	87.54
JLCX1	78.76	77.30	76.14	63.69	62.71	84.86	83.12	89.21
JLCX2	79.38	77.92	76.74	64.22	63.20	85.55	83.53	90.28
SG1	77.92	72.41	71.04	58.26	57.10	77.95	87.70	73.46

download these results as csv

JayChou29

Algorithm	MirexRoot	MirexMajMin	MirexMajMinBass	MirexSevenths	MirexSeventhsBass	MeanSeg	UnderSeg	OverSeg
CM1	72.82	72.15	65.55	54.46	49.05	86.58	86.94	86.84
FK2	81.42	80.48	69.66	51.77	43.18	89.08	87.67	90.87
JLCX1	84.41	84.12	81.42	63.14	61.08	92.59	91.82	93.46
JLCX2	84.77	84.49	81.75	63.38	61.31	93.16	92.26	94.16
SG1	80.59	75.65	72.28	56.43	53.80	82.51	91.49	75.30

download these results as csv

RobbieWilliams

Algorithm	MirexRoot	MirexMajMin	MirexMajMinBass	MirexSevenths	MirexSeventhsBass	MeanSeg	UnderSeg	OverSeg
CM1	81.94	78.29	76.09	57.97	55.94	87.95	89.00	87.39
FK2	89.38	88.12	85.06	83.38	80.68	90.96	89.52	92.84
JLCX1	87.14	85.33	83.31	75.91	74.10	90.95	89.30	93.23
JLCX2	87.25	85.45	83.40	76.01	74.17	91.03	89.04	93.73
SG1	84.87	80.28	78.58	73.78	72.16	85.24	90.85	80.75

download these results as csv

RWC-Popular

Algorithm	MirexRoot	MirexMajMin	MirexMajMinBass	MirexSevenths	MirexSeventhsBass	MeanSeg	UnderSeg	OverSeg
CM1	79.12	77.95	74.30	63.33	59.89	88.62	88.11	89.67
FK2	87.20	86.58	80.92	58.04	52.78	89.94	87.94	92.37
JLCX1	87.11	86.71	84.93	73.78	72.17	91.15	88.77	93.99
JLCX2	86.90	86.53	84.76	73.51	71.88	91.07	88.48	94.16
SG1	82.22	78.06	76.06	63.65	61.86	82.62	90.66	76.44

download these results as csv

USPOP2002Chords

Algorithm	MirexRoot	MirexMajMin	MirexMajMinBass	MirexSevenths	MirexSeventhsBass	MeanSeg	UnderSeg	OverSeg
CM1	78.41	76.24	72.74	59.84	56.67	85.81	86.81	86.02
FK2	82.34	80.72	76.04	64.50	60.30	86.59	85.49	89.17
JLCX1	85.66	84.89	82.95	75.08	73.35	89.57	88.22	91.88
JLCX2	85.34	84.67	82.72	74.86	73.12	89.35	87.72	92.00
SG1	81.67	77.38	75.16	65.69	63.68	83.46	90.37	78.59

download these results as csv

Detailed Results

More details about the performance of the algorithms, including per-song performance and supplementary statistics, are available from ?.

Algorithmic Output

The raw output of the algorithms are available in ?. They can be used to experiment with alternative evaluation measures and statistics.