EURASIP Journal on Audio, Speech, and Music Processing

Table 3 Identification accuracy in four noisy conditions (white, pink, factory, and f16) for TIMIT dataset.

From: Auditory Sparse Representation for Robust Speaker Recognition Based on Tensor Structure

(%)	SNR	ANTCC	MFCC	LPCC	RASTA-PLP
White	0 dB	2.9	1.43	2.38	2.38
	5 dB	3.81	2.38	2.86	5.24
	10 dB	29.52	3.33	6.19	15.71
	15d B	64.29	11.43	12.86	39.52
Pink	0 dB	2.43	1.43	3.33	1.43
	5 dB	13.81	1.9	3.81	5.24
	10 d	50.95	8.57	8.1	27.14
	15 d	78.57	30	32.86	60.95
Factory	0 dB	2.43	1.43	2.76	1.43
	5 dB	12.86	3.33	10.48	10
	10 d	49.52	21.9	34.29	46.67
	15 d	78.1	70	73.81	74.76
F16	0 dB	2.9	2.86	2.33	1.43
	5 dB	15.24	7.14	14.76	8.1
	10 d	47.14	24.76	28.57	34.76
	15 d	77.62	57.14	67.62	60.48

Back to article page