2009:Audio Music Mood Classification Results

From MIREX Wiki
Revision as of 12:31, 23 October 2009 by Muntash1 (talk | contribs) (Team ID)

Introduction

These are the results for the 2009 running of the Audio Music Mood Classification task. For background information about this task set please refer to the Audio Music Mood Classification page. The data was created by Xiao Hu and consists of 600 files organized into 5 mood "clusters".

Mood Clusters

The 5 mood clusters were derived from the AMG mood repository.

   * Cluster_1: passionate, rousing, confident,boisterous, rowdy
   * Cluster_2: rollicking, cheerful, fun, sweet, amiable/good natured
   * Cluster_3: literate, poignant, wistful, bittersweet, autumnal, brooding
   * Cluster_4: humorous, silly, campy, quirky, whimsical, witty, wry
   * Cluster_5: aggressive, fiery,tense/anxious, intense, volatile,visceral 

For more information on the clusters, please see

Hu, Xiao and J. Stephen Downie (2007) Exploring mood metadata: Relationships with genre, artist and usage metadata, In the 8th International Conference on Music Information Retrieval (ISMIR 2007), Vienna, September 23-27, 2007.

Data

There are 600 audio clips with 120 in each mood cluster. Each clip belongs to only one mood cluster. The clips were chosen from the APM audio set .

The mood cluster labels of the clips were firstly suggested by their metadata provided by APM and then decided by human validations using the Evalutron6000

Each mood cluster covers a variety of genres: each category covers about 7 major genres (with 20-30 tracks each) and a few minor genres, and the distribution among major genres within each category is made as even as possible.

Audio format: 30 second clips, 22.05kHz, mono, 16bit, WAV files; The data were evenly split into 3 folds.

For more information on the dataset and evaluation methods, please see

X. Hu, J. S. Downie, C. Laurier, M. Bay, A.Ehmann (2008) The 2007 MIREX Audio Mood Classification Task: Lessons Learned, In the 9th International Symposium on Music Information Retrieval (ISMIR 2008), Philadelphia, Sept. 2008



General Legend

Team ID

ANO= Anonymous
BP1= Juan José Burred, Geoffroy Peeters (file)
BP2 = Juan José Burred, Geoffroy Peeters (tw)
CL1 = Chuan Cao, Ming Li
CL2 = Chuan Cao, Ming Li
FCY1 = Tao Feng, XiaoOu Chen, DeShun Yang
FCY2 = Tao Feng, XiaoOu Chen, DeShun Yang
GP = Geoffroy Peeters
GT1 = George Tzanetakis (mono)
GT2 = George Tzanetakis (stereo)
GLR1 = A. Grecu, T. Lidy, A. Rauber (full)
GLR2 = A. Grecu, T. Lidy, A. Rauber (template)
HNOS1 = Takashi Hasegawa, Takuya Nishimoto, Nobutaka Ono, Shigeki Sagayama (tcca)
HNOS2 = Takashi Hasegawa, Takuya Nishimoto, Nobutaka Ono, Shigeki Sagayama (tcck)
HNOS3 = Takashi Hasegawa, Takuya Nishimoto, Nobutaka Ono, Shigeki Sagayama (tccl)
HNOS4 = Takashi Hasegawa, Takuya Nishimoto, Nobutaka Ono, Shigeki Sagayama (tcpk)
HW1 = Huaxin Wang
HW2 = Huaxin Wang
VA1 = T. Lidy, A. Grecu, A. Rauber, A. Pertusa, P. J. Ponce de Léon, J. M. Iñesta (WMV)
VA2 = T. Lidy, A. Grecu, A. Rauber, A. Pertusa, P. J. Ponce de Léon, J. M. Iñesta (BWWV)
LZG = Yi Liu, Tao Zheng, Yue Gao (RUC_1)
RK1 = Preeti Rao, Sujeet Kini
RK2 = Preeti Rao, Sujeet Kini
SS = Klaus Seyerlehner, Markus Schedl
TAOS= Emiru Tsunoo, Taichi Akase, Nobutaka Ono, Shigeki Sagayama
MTG1 = N. Wack, E. Guaus, C. Laurier, O. Meyers, R. Marxer, D. Bogdanov, J. Serrà, P. Herrera (false, rca)
MTG2 = N. Wack, E. Guaus, C. Laurier, O. Meyers, R. Marxer, D. Bogdanov, J. Serrà, P. Herrera (true, rca)
MTG3 = N. Wack, E. Guaus, C. Laurier, O. Meyers, R. Marxer, D. Bogdanov, J. Serrà, P. Herrera (false, simca)
MTG4 = N. Wack, E. Guaus, C. Laurier, O. Meyers, R. Marxer, D. Bogdanov, J. Serrà, P. Herrera (true, simca)
MTG5 = N. Wack, E. Guaus, C. Laurier, O. Meyers, R. Marxer, D. Bogdanov, J. Serrà, P. Herrera (false, svm)
MTG6 = N. Wack, E. Guaus, C. Laurier, O. Meyers, R. Marxer, D. Bogdanov, J. Serrà, P. Herrera (true, svm)
XLZZG = Jieping Xu, Yi Liu, Tao Zheng, Chao Zhen, Yue Gao (RUC_1)
XZZ = JiePing Xu, Chao Zhen, Tao Zheng (RUC_2)

Overall Summary Results

Raw Classification Accuracy Averaged Over Three Train/Test Folds

file /nema-raid/www/mirex/results/audiomood/summary_audiomood.csv not found

Accuracy Across Folds

file /nema-raid/www/mirex/results/audiomood/audiomood_Accuracy.csv not found

Accuracy Across Categories

file /nema-raid/www/mirex/results/audiomood/audiomood_Accuracy_Per_Class.csv not found

Friedman's Tests for Significant Differences

Classes vs. System Tukey-Kramer HSD Multi-Comparisons

The Friedman test was run in MATLAB against the average accuracy for each class. The Tukey-Kramer HSD multi-comparison data below was generated using the following MATLAB instruction. Command:
[c, m, h, gnames] = multicompare(stats, 'ctype', 'tukey-kramer', 'estimate', 'friedman', 'alpha', 0.05);

file /nema-raid/www/mirex/results/audiomood/audiomood_Accuracy_Per_Class.friedman.tukeyKramerHSD.csv not found

https://music-ir.org/mirex/2009/results/audiomood/small.audiomood_Accuracy_Per_Class.friedman.tukeyKramerHSD.png

Folds vs. Systems Tukey-Kramer HSD Multi-Comparison

The Friedman test was run in MATLAB against the accuracy for each fold. The Tukey-Kramer HSD multi-comparison data below was generated using the following MATLAB instruction. Command:
[c, m, h, gnames] = multicompare(stats, 'ctype', 'tukey-kramer', 'estimate', 'friedman', 'alpha', 0.05);

file /nema-raid/www/mirex/results/audiomood/audiomood_Accuracy.friedman.tukeyKramerHSD.csv not found

https://music-ir.org/mirex/2009/results/audiomood/small.audiomood_Accuracy.friedman.tukeyKramerHSD.png

Results By Algorithm

(.tgz)

ANO= Anonymous
BP1= Juan José Burred, Geoffroy Peeters (file)
BP2 = Juan José Burred, Geoffroy Peeters (tw)
CL1 = Chuan Cao, Ming Li
CL2 = Chuan Cao, Ming Li
FCY1 = Tao Feng, XiaoOu Chen, DeShun Yang
FCY2 = Tao Feng, XiaoOu Chen, DeShun Yang
GP = Geoffroy Peeters
GT1 = George Tzanetakis (mono)
GT2 = George Tzanetakis (stereo)
GLR1 = A. Grecu, T. Lidy, A. Rauber (full)
GLR2 = A. Grecu, T. Lidy, A. Rauber (template)
HNOS1 = Takashi Hasegawa, Takuya Nishimoto, Nobutaka Ono, Shigeki Sagayama (tcca)
HNOS2 = Takashi Hasegawa, Takuya Nishimoto, Nobutaka Ono, Shigeki Sagayama (tcck)
HNOS3 = Takashi Hasegawa, Takuya Nishimoto, Nobutaka Ono, Shigeki Sagayama (tccl)
HNOS4 = Takashi Hasegawa, Takuya Nishimoto, Nobutaka Ono, Shigeki Sagayama (tcpk)
HW1 = Huaxin Wang
HW2 = Huaxin Wang
VA1 = T. Lidy, A. Grecu, A. Rauber, A. Pertusa, P. J. Ponce de Léon, J. M. Iñesta (WMV)
VA2 = T. Lidy, A. Grecu, A. Rauber, A. Pertusa, P. J. Ponce de Léon, J. M. Iñesta (BWWV)
LZG = Yi Liu, Tao Zheng, Yue Gao (RUC_1)
RK1 = Preeti Rao, Sujeet Kini
RK2 = Preeti Rao, Sujeet Kini
SS = Klaus Seyerlehner, Markus Schedl
TAOS= Emiru Tsunoo, Taichi Akase, Nobutaka Ono, Shigeki Sagayama
MTG1 = N. Wack, E. Guaus, C. Laurier, O. Meyers, R. Marxer, D. Bogdanov, J. Serrà, P. Herrera (false, rca)
MTG2 = N. Wack, E. Guaus, C. Laurier, O. Meyers, R. Marxer, D. Bogdanov, J. Serrà, P. Herrera (true, rca)
MTG3 = N. Wack, E. Guaus, C. Laurier, O. Meyers, R. Marxer, D. Bogdanov, J. Serrà, P. Herrera (false, simca)
MTG4 = N. Wack, E. Guaus, C. Laurier, O. Meyers, R. Marxer, D. Bogdanov, J. Serrà, P. Herrera (true, simca)
MTG5 = N. Wack, E. Guaus, C. Laurier, O. Meyers, R. Marxer, D. Bogdanov, J. Serrà, P. Herrera (false, svm)
MTG6 = N. Wack, E. Guaus, C. Laurier, O. Meyers, R. Marxer, D. Bogdanov, J. Serrà, P. Herrera (true, svm)
XLZZG = Jieping Xu, Yi Liu, Tao Zheng, Chao Zhen, Yue Gao (RUC_1)
XZZ = JiePing Xu, Chao Zhen, Tao Zheng (RUC_2)

Run Times

file /nema-raid/www/mirex/results/mood.runtime.csv not found TBA