Skip to main content

Advertisement

Table 2 Accuracy rates (%) for the different extraction techniques

From: Auditory processing-based features for improving speech recognition in adverse acoustic conditions

Channel GFCC-MS GFCC-MS-NL GFCC-NL
1 87.3 89.9 89.6
2 80.7 83.4 81.1
3 62.6 72.8 70.9
4 56.0 69.2 65.6
5 64.5 71.6 69.4
6 62.6 73.2 70.1
7 62.0 66.9 68
8 69.7 74.4 71
9 61.7 63.8 62.4
10 48.2 52.3 51.9
11 44.5 51.2 50.1
12 43.0 50.8 48.4
13 47.2 54.3 51.6
14 46.8 50.4 50.2
Average 59.8 64.3 66.0