COVID-19 detection on Cambridge COVID-19 Sound database
86AccuracyHuBERT
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| HuBERTInput=Encoder features2024.02 | 86 | 80 | 89 | 82 | 90 | 0.93 | |
| Optimized MFCCModality=Only Speech2024.08 | 81.1 | — | — | — | — | — | |
| SVMInput=MFCC vector, kernel=rbf, C=1, gamma=0.0012024.02 | 81 | 87 | 80 | 54 | 96 | — | |
| LSTMInput=MFCC vector, loss=mean absolute error, optimizer=adam, activation=sigmoid2024.02 | 81 | 63 | 90 | 77 | 83 | 0.8 | |
| CNNInput=Mel-spectrogram images, input shape=(150,150,3), loss=binary crossentropy, optimizer=adam, activation=softmax2024.02 | 78 | 65 | 85 | 70 | 82 | 0.84 | |
| LRInput=MFCC vector2024.02 | 71 | 62 | 73 | 39 | 87 | 0.72 | |
| CNNInput=MFCC images, input shape=(150,150,3), loss=binary crossentropy, optimizer=adam, activation=softmax2024.02 | 59 | 38 | 69 | 33 | 72 | 0.54 |