EURASIP Journal on Audio, Speech, and Music Processing

Table 2 Measures compared with various VAD models trained on DEC-1100 dataset and tested on DEC-based human annotated test set

From: Language agnostic missing subtitle detection

Model name	Accuracy	AUC	Precision	Recall	F-score
CLDNN	0.852	0.915	0.877	0.852	0.854
CNNTD	0.866	0.947	0.876	0.866	0.867
GRU	0.871	0.951	0.887	0.871	0.872
STNET	0.863	0.940	0.863	0.863	0.861
TCN	0.875	0.900	0.887	0.875	0.876
WebRTC	0.615	-	0.757	0.615	0.597

Back to article page