Table 16 Accuracy of the proposed sconn model compared with those of other models on the TESS dataset.
From: Stacked convolutional neural network for emotion recognition using multi feature speech analysis
Author’s Name | Features | Methodology | Accuracy |
|---|---|---|---|
Dolka et al.14 | MFCCS | Artificial Neural Network | 99.52% |
Tellai et al. 25 | Mel Spectrogram, Mel Frequency Cepstral Coefficients | CNN-Transformer | 99.42% |
Choudhary et al.49 | MFCC | CNN | 97.10% |
Huang et al. 50 | MFCC | 2-D CNN | 85.00% |
This Work | Mel Spectrogram | SCoNN | 99.68% |
MFCC | 99.93% | ||
Combined | 99.93% |
