Ensemble Learning by High-Dimensional Acoustic Features for Emotion Recognition from Speech Audio Signal

Table 1

The mean and deviation of the assessment metric values depicted from multifold cross-validation.

Average of 10-fold result and deviations
Metrics	EL-HDAF	HAF	SBL-DF
Precision	0.944679 ± 0.032751	0.894171 ± 0.058389	0.880646 ± 0.065648
Sensitivity recall	0.954865 ± 0.012566	0.908286 ± 0.010922	0.896128 ± 0.017381
Specificity	0.954897 ± 0.005458	0.907739 ± 0.002584	0.893572 ± 0.010829
F-score	0.951408 ± 0.018341	0.899709 ± 0.031556	0.886927 ± 0.034786
Decision accuracy	0.948429 ± 0.030393	0.894171 ± 0.058389	0.880646 ± 0.065648