Table 9 Feature selection algorithm comparison (Bat algorithm: best accuracy/time trade-off).

From: A deep learning framework for gender sensitive speech emotion recognition based on MFCC feature selection and SHAP analysis

Training algorithm

Training and test time

Male detection (%)

Female detection (%)

Mean detection

Bat 1

12 h

92

87

89.5

Bat 2

22 h

92

90

91

Bat 4

36 h

93

91

92

Tlbo 1

16 h

86

85

85.5

Tlbo 2

24 h

87

85

86

Tlbo 4

72 h

88

86

87

GSA

24 h

69.5

68

68.7

GA 1

12 h

82

81

81.5

GA 2

20 h

84

82

83

Fisher

12 h

75.5

74

74.8

Dempster-Shafer

18 h

80

79

80.5