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

EURASIP Journal on Audio, Speech, and Music Processing

Table 9 Results of the ALBAYZIN QbE STD 2014 evaluation on train/dev data

System ID	MTWV	ATWV	p(FA)	p(Miss)
Fusion (SGMM) + Posteriorgram	0.3023	0.3023	0.00009	0.607
Fusion (SGMM)	0.2957	0.2957	0.00009	0.616
Fusion (DNN) + Posteriorgram	0.3411	0.3388	0.00007	0.590
Fusion (DNN)	0.3394	0.3375	0.00007	0.593
Text-based SGMM STD	0.5639	0.5639	0.00008	0.358
Text-based DNN STD	0.6112	0.6062	0.00006	0.327