Skip to main content

Advertisement

Table 5 Accuracy performance of emotion recognition [%] without selection (–) in comparison with two different selection methods (FMS and SFS) and extraction (S-PCA) methods using SVM algorithm as a classifier

From: Efficiency of chosen speech descriptors in relation to emotion recognition

  Acted speech Natural speech
  SFS FMS S-PCA SFS FMS S-PCA
f 0 19.66 19.66 19.66 34.17 26.98 26.98 26.98 52.55
f 1f 3 27.5 27.5 25.31 40.92 16.877 17.29 27.5 57.38
Energy 29.95 29.95 29.95 35.86 46.87 46.87 47.72 53.97
LPC 29.91 32.49 32.06 32.06 60.51 65.35 60.79 64.20
BFCC 40.92 47.67 46.83 56.11 59.4 62.53 59.40 81.7
MFCC 40.5 45.14 41.35 54.96 58.31 63.48 58.31 77.92
HFCC 39.66 46.83 40.08 57.38 55.17 56.13 55.17 77.79
PLP 22.47 22.47 22.47 46.83 21.51 22.36 22.47 73.56
RPLP 43.88 44.72 44.72 57.80 54.76 59.12 54.76 70.31
RASTA 36.28 39.66 36.28 51.9 50 60.04 62.53 73.84