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

EURASIP Journal on Audio, Speech, and Music Processing

Table 11 Results of the QbE STD systems based on query speaking speed on test data

System ID	Speaking	MTWV	ATWV	p(FA)	p(miss)
	speed
System 1	Slow	0.0655	0.0543	0.00000	0.935
	Medium	0.0560	0.0445	0.00001	0.937
	Fast	0.0418	-0.0678	0.00000	0.958
System 2a	Slow	0.0083	0.0054	0.00003	0.957
	Medium	0.0083	0.0023	0.00000	0.992
	Fast	0.0090	0.0015	0.00001	0.984
System 2b	Slow	0.0121	0.0103	0.00000	0.988
	Medium	0.0083	0.0023	0.00000	0.992
	Fast	0.0090	0.0015	0.00001	0.984
System 3a	Slow	0.0050	-1.1229	0.00000	0.995
	Medium	0.0000	-2.9868	0.00000	1.000
	Fast	0.0000	-2.2972	0.00000	1.000
System 3b	Slow	0.0000	-0.0257	0.00000	1.000
	Medium	0.0000	-0.1072	0.00000	1.000
	Fast	0.0000	-0.0684	0.00000	1.000
System 4a	Slow	0.0000	-0.5006	0.00000	1.000
	Medium	0.0000	-0.7150	0.00000	1.000
	Fast	0.0000	-0.7088	0.00000	1.000
System 4b	Slow	0.0775	0.0775	0.00004	0.881
	Medium	0.0000	-0.0416	0.00000	1.000
	Fast	0.0330	0.0330	0.00007	0.894