Nonlinear Dynamic Feature Extraction Based on Phase Space Reconstruction for the Classification of Speech and Emotion

Table 3

Emotional speech recognition using different phase space parameters (%).

The choice of and	Happy	Sadness	Neutral	Anger	Fear	Average
Literature [20] sets and	41.7	90.0	73.1	88.9	73.9	73.3
The paper sets and	75.0	85.0	73.1	81.5	65.2	75.8