Table 11: Performance Comparison of Various Combination of Facial Features with Audio Feature for Proposed BPN Approach.

SNRsType
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 dB21.6726.6733.33
5 dB29.3340.2543.67
10 dB45.3356.2559.67
15 dB73.2579.0083.25
20 dB84.6790.6792.33
25 dB90.0093.0096.00
30 dB95.0096.3398.67