MIREX 2015: Audio Train/Test: Classical Composer Identification - MIREX08 Dataset


Accuracy per Class


Legend

CYC1
CK
CYC2
CK2
CYC3
CK3
LK2
Audio classification
LK3
Audio Classification
LP1
Structural Analysis for Composer Identification
LP2
Structural Analysis for Composer Identification (DBN)
LP3
Structural Analysis for Composer Identification (NN + Means MFCC)
LP4
Structural Analysis for Composer Identification (NN + Means MFCC + chord analysis)


Accuracy per Class    [top]

bach
77.7778
78.1746
76.1905
86.1111
68.6508
48.4127
65.4762
48.8095
38.8889
beethoven
51.1905
48.8095
53.9683
63.8889
48.8095
9.1270
4.3651
48.0159
17.4603
brahms
55.5556
55.1587
52.3810
65.8730
41.2698
23.4127
13.0952
26.1905
24.6032
chopin
78.9683
79.7619
74.2063
85.3175
80.9524
71.0317
81.3492
67.8571
74.6032
dvorak
58.3333
57.9365
48.4127
60.3175
46.4286
11.5079
20.6349
31.7460
34.5238
handel
59.9206
61.1111
58.3333
73.0159
53.9683
32.9365
26.1905
34.1270
36.1111
haydn
63.4921
62.3016
65.8730
71.8254
60.3175
0.0000
0.0000
0.0000
0.0000
mendelssohnn
43.6508
41.2698
44.0476
59.5238
42.8571
24.6032
29.3651
34.1270
40.0794
mozart
54.3651
52.7778
38.8889
53.5714
37.6984
10.3175
5.1587
12.3016
9.9206
schubert
41.6667
42.4603
39.6825
59.9206
35.7143
18.6508
19.8413
34.5238
28.1746
vivaldi
74.2063
71.4286
71.0317
76.9841
56.7460
45.2381
19.4444
42.8571
56.7460