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

Figure 2

Fourfold cross-validation determined the positive prediction rate (precision).