A summary of parameters and performances of the used models
| Model | Parameters | Accuracy | Sensitivity | Specificity | PPV | NPV |
|---|---|---|---|---|---|---|
| LR | Input = MFCC vector | 0.71 ± 0.04 | 0.62 ± 0.11 | 0.73 ± 0.03 | 0.39 ± 0.12 | 0.87 ± 0.05 |
| SVM | Input = MFCC vector, kernel = rbf, C = 1, gamma = 0.001 | 0.81 ± 0.04 | 0.87 ± 0.01 | 0.80 ± 0.03 | 0.54 ± 0.08 | 0.96 ± 0.03 |
| CNN | Input = MFCC images, input shape = (150,150,3), loss = binary crossentropy, optimizer = adam, activation = softmax | 0.59 ± 0.11 | 0.38 ± 0.32 | 0.69 ± 0.31 | 0.33 ± 0.20 | 0.72 ± 0.11 |
| LSTM | Input = MFCC vector, loss = mean absolute error, optimizer = adam, activation = sigmoid | 0.81 ± 0.03 | 0.63 ± 0.06 | 0.90 ± 0.04 | 0.77 ± 0.08 | 0.83 ± 0.03 |
| CNN | Input = Mel-spectrogram images, input shape = (150,150,3), loss = binary crossentropy, optimizer = adam, activation = softmax | 0.78 ± 0.03 | 0.65 ± 0.12 | 0.85 ± 0.04 | 0.70 ± 0.04 | 0.82 ± 0.04 |
| HuBERT | Input = Encoder features | 0.86 ± 0.03 | 0.80 ± 0.09 | 0.89 ± 0.07 | 0.82 ± 0.08 | 0.90 ± 0.04 |