EURASIP Journal on Audio, Speech, and Music Processing

Table 8 Overall system results of the ALBAYZIN 2018 QbE STD evaluation on development and test data (average of the results on the two development and three test corpora)

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

	Development		Test
System ID	MTWV	ATWV	MTWV	ATWV
A-Hybrid DTW+LVCSR	0.5085	0.4717	0.3260	0.2084
B-Fusion DTW	0.4955	0.4863	0.3082	− 0.4185
C-PhonePost DTW	0.4558	0.4412	0.2891	− 0.4434
D-LVCSR	0.2962	0.2779	0.2080	0.1172
E-DTW	0.1094	0.1070	0.0995	0.0948
F-Combined DTW	0.3758	0.3406	0.1657	0.1413
G-Super-BNF DTW	0.3776	0.3668	0.1460	− 0.0489
H-Multilingual-BNF DTW	0.3379	0.3266	0.1570	− 0.2018
I-Monoph.-BNF DTW	0.3621	0.3342	0.1689	0.0422
J-Triph.-BNF DTW	0.3527	0.3382	0.1628	− 0.0264
K-Text STD	0.6817	0.6480	0.4427	0.2012

Back to article page