EURASIP Journal on Audio, Speech, and Music Processing

Table 6 SER, error per class and average class error for the BLSTM₁PoolBLSTM₂ RNN classifier on the test partition for different pooling factors (N) and average pooling

From: Multiclass audio segmentation based on recurrent neural networks for broadcast domain data

N	SER	Class error (%)				Avg
		mu	sp	sm	sn
5	15.49	15.57	28.71	24.63	30.71	24.90
10	15.47	15.55	29.16	24.34	30.51	24.89
25	15.51	16.85	27.17	26.23	30.06	25.08
50	15.53	16.40	28.60	24.92	30.85	25.19
75	15.54	16.87	27.97	25.55	29.88	25.07
100	15.49	18.22	26.77	24.91	29.80	24.92
No pool	15.91	16.28	28.82	26.32	31.94	25.84

Back to article page