Table 1 HIT-FA results for feature groups in the noise mismatching and matching condition
From: Research on monaural speech segregation based on feature selection
Feature and feature group | Mismatching | Matching |
|---|---|---|
AMS + GF | 65.90% | 77.80% |
AMS + GF + MRCG | 77.12% | 79.41% |
AMS + MFCC | 69.65% | 76.48% |
AMS + MFCC + GF | 74.02% | 79.20% |
AMS + MFCC + GF + MRCG | 77.57% | 79.85% |
AMS + MFCC + MRCG | 77.31% | 79.44% |
AMS + MRCG | 76.78% | 78.73% |
GF + MRCG | 76.84% | 78.87% |
MFCC + GF | 72.64% | 78.25% |
MFCC + GF + MRCG | 77.04% | 79.45% |
MFCC + MRCG | 76.71% | 79.12% |