Query-by-Example Spoken Term Detection ALBAYZIN 2012 evaluation: overview, systems, results, and discussion

EURASIP Journal on Audio, Speech, and Music Processing

Table 8 Results of the QbE STD systems for different time window tolerance on test data

System ID	Time window	MTWV	ATWV	p(FA)	p(miss)
	(sec)
System 1	0.5	0.0436	0.0122	0.00000	0.952
	5	0.0436	0.0194	0.00000	0.952
	10	0.0436	0.0194	0.00000	0.952
	30	0.0436	0.0194	0.00000	0.952
System 2a	0.5	0.0055	0.0031	0.00001	0.983
	5	0.0055	0.0031	0.00001	0.983
	10	0.0098	0.0075	0.00001	0.981
	30	0.0098	0.0075	0.00001	0.981
System 2b	0.5	0.0075	0.0047	0.00000	0.990
	5	0.0075	0.0047	0.00000	0.990
	10	0.0075	0.0047	0.00000	0.990
	30	0.0075	0.0047	0.00000	0.990
System 3a	0.5	0.0000	-2.1471	0.00000	1.000
	5	0.0000	-2.0841	0.00000	1.000
	10	0.0000	-2.0560	0.00000	1.000
	30	0.0000	-1.8739	0.00000	1.000
System 3b	0.5	0.0000	-0.0678	0.00000	1.000
	5	0.0000	-0.0644	0.00000	1.000
	10	0.0000	-0.0610	0.00000	1.000
	30	0.0000	-0.0585	0.00000	1.000
System 4a	0.5	0.0000	-0.6416	0.00000	1.000
	5	0.0000	-0.5746	0.00000	1.000
	10	0.0000	-0.5522	0.00000	1.000
	30	0.0000	-0.4952	0.00000	1.000
System 4b	0.5	0.0238	0.0217	0.00009	0.884
	5	0.0432	0.0361	0.00008	0.874
	10	0.0469	0.0398	0.00008	0.873
	30	0.0556	0.0438	0.00008	0.869