EURASIP Journal on Audio, Speech, and Music Processing

Table 1 Performance of baseline language recognition system

From: Homogenous ensemble phonotactic language recognition based on SVM supervector reconstruction

	SVM supervector	30 s		10 s		3 s
	dimension	EER	Cavg	EER	Cavg	EER	Cavg
HU	195112	2.44	2.39	7.54	7.38	24.23	23.98
RU	117649	2.26	2.06	6.23	6.07	20.53	20.38
CZ	74088	3.39	3.31	10.13	10.04	28.73	28.35

NIST LRE 2009 (EER and Cavg in percent).

Back to article page