BPN Based Likelihood Ratio Score Fusion for Audio-Visual Speaker Identification in Response to Noise

Table 11

Performance Comparison of Various Combination of Facial Features with Audio Feature for Proposed BPN Approach.

SNRs	Type
	Appearance and ΔMFCC based feature with BPN approach (in %)	Shape and ΔMFCC based feature with BPN approach (in %)	Appearance, shape and ΔMFCC based feature with BPN approach (in %)
0 dB	21.67	26.67	33.33
5 dB	29.33	40.25	43.67
10 dB	45.33	56.25	59.67
15 dB	73.25	79.00	83.25
20 dB	84.67	90.67	92.33
25 dB	90.00	93.00	96.00
30 dB	95.00	96.33	98.67