Table 4 Accuracy of deep learning models in audio evaluation on GTZAN dataset. All models are trained and tested under the same conditions (hardware) without using pre trained models.
From: Dense dynamic convolutional network for Bel canto vocal technique assessment
