EURASIP Journal on Audio, Speech, and Music Processing

Table 21 System results of the ALBAYZIN 2018 QbE STD evaluation on RTVE test data for in-vocabulary (INV) and out-of-vocabulary (OOV) queries

From: Search on speech from spoken queries: the Multi-domain International ALBAYZIN 2018 Query-by-Example Spoken Term Detection Evaluation

	INV				OOV
System ID	MTWV	ATWV	p(FA)	p(Miss)	MTWV	ATWV	p(FA)	p(Miss)
A-Hybrid DTW+LVCSR	0.6948	0.6449	0.00007	0.233	0.3401	0.2590	0.00005	0.608
B-Fusion DTW	0.4525	0.3969	0.00004	0.507	0.3879	0.3388	0.00001	0.603
C-PhonePost DTW	0.4277	0.3540	0.00002	0.555	0.3837	0.2814	0.00002	0.595
D-LVCSR	0.5527	0.4920	0.00008	0.369	0.1674	0.0061	0.00003	0.800
E-DTW	0.0000	− 0.0175	0.00000	1.000	0.0000	0.0000	0.00000	1.000
F-Combined DTW	0.3303	0.3091	0.00004	0.633	0.3072	0.2927	0.00010	0.591
G-Super-BNF DTW	0.3057	0.3033	0.00003	0.661	0.2932	0.2865	0.00003	0.672
H-Multilingual-BNF DTW	0.2970	0.2918	0.00004	0.661	0.2812	0.2663	0.00003	0.689
I-Monoph.-BNF DTW	0.3427	0.3424	0.00005	0.606	0.2737	0.2461	0.00009	0.637
J-Triph.-BNF DTW	0.3092	0.2956	0.00003	0.657	0.3218	0.2979	0.00004	0.638
K-Text STD	0.7896	0.7733	0.00003	0.179	0.4391	0.2344	0.00011	0.451

Back to article page