Table 2 Accuracy for various network models trained and tested using the Jamendo dataset
From: Comparative study of singing voice detection based on deep neural networks and ensemble learning
Network model | Average accuracy (10 trials) (%) |
|---|---|
MCNN | 88.2 |
SCNN | 91.8 |
ECNN | 77.1 |
SLSTM | 87.7 |
CLSTM | 90.4 |
Capsule net | 82.5 |