Table 12 Sensitivity and specificity of the sconn model for each emotion class using the mel Spectrogram, MFCC, and mel Spectrogram + MFCC features of the SAVEE dataset.
From: Stacked convolutional neural network for emotion recognition using multi feature speech analysis
Emotion Labels | Mel Spectrogram | MFCC | Mel Spectrogram + MFCC | |||
|---|---|---|---|---|---|---|
Sensitivity | Specificity | Sensitivity | Specificity | Sensitivity | Specificity | |
angry | 100.00% | 99.16% | 92.19% | 98.03% | 93.65% | 100.00% |
disgust | 93.85% | 99.44% | 90.77% | 100.00% | 92.96% | 99.27% |
fear | 96.61% | 99.45% | 91.53% | 98.06% | 95.16% | 100.00% |
happy | 96.55% | 98.62% | 84.48% | 97.79% | 91.07% | 99.29% |
neutral | 92.65% | 99.15% | 97.06% | 99.72% | 95.45% | 98.92% |
sad | 92.16% | 98.37% | 94.12% | 99.19% | 95.83% | 98.61% |
surprised | 90.91% | 99.73% | 89.09% | 97.26% | 100.00% | 98.05% |
