Comparison of ALBAYZIN query-by-example spoken term detection 2012 and 2014 evaluations

EURASIP Journal on Audio, Speech, and Music Processing

Table 12 Results of the ALBAYZIN QbE STD 2012 and 2014 evaluations on the ALBAYZIN QbE STD 2012 train/dev data

System ID	MTWV	ATWV	p(FA)	p(Miss)
Fusion (SGMM) + Posteriorgram	0.2850	0.2850	0.00011	0.610
Fusion (SGMM)	0.2824	0.2803	0.00010	0.619
Fusion (DNN) + Posteriorgram	0.3572	0.3514	0.00007	0.578
Fusion (DNN)	0.3579	0.3571	0.00006	0.580
DTW-Zero	0.0455	0.0455	0.00002	0.930
P1B-STDa	0.0128	0.0128	0.00000	0.986
P1B-STDb	0.0092	0.0092	0.00000	0.990
P1L-STD	0.0000	0.0000	0.00000	1.000
DTW-Spanish	0.0612	0.0612	0.00005	0.893
Text-based SGMM STD	0.6866	0.6866	0.00009	0.226
Text-based DNN STD	0.7440	0.7398	0.00006	0.192