Skip to main content

Table 4 Classification results with matched training/testing using different acoustic features

From: Biomimetic spectro-temporal features for music instrument recognition in isolated notes and solo phrases

 

Rates (22dim)

Scales (11dim)

Freq (128dim)

RateScale (242dim)

ScaleFreq (420dim)

RateFreq (420dim)

RateScaleFreq (420dim)

RWC

73.3 ± 1.2 %

57.5 ± 0.9 %

93.5 ± 0.7 %

93.8 ± 0.8 %

97.7 ± 0.4 %

97.5 ± 0.5 %

98.5 ± 0.2 %

Notes

69.6 ± 1.9 %

67.0 ± 1.9 %

93.0 ± 0.9 %

90.0 ± 0.6 %

96.0 ± 0.8 %

96.1 ± 1.1 %

97.7 ± 0.5 %