Table 1 SNR Results. (Mixture: Original degraded speech; Hu-Wang: Hu and Wang model; Proposed: Proposed model; TP Hu-Wang: true pitch-based Hu and Wang model; TP proposed: true pitch-based proposed model; IBM: Ideal binary mask)
From: Monaural Voiced Speech Segregation Based on Dynamic Harmonic Function
| Â | N0 | N1 | N2 | N3 | N4 | N5 | N6 | N7 | N8 | N9 | Avg |
|---|---|---|---|---|---|---|---|---|---|---|---|
Mixture | −7.42 | −8.27 | 5.62 | 0.80 | 0.68 | −10.00 | −1.62 | 3.85 | 9.53 | 2.75 | −0.41 |
Hu-Wang | 16.01 | 5.59 | 14.27 | 5.83 | 8.25 | 14.35 | 15.53 | 10.46 | 14.06 | 6.88 | 11.12 |
Proposed | 17.95 | 6.32 | 17.76 | 6.51 | 9.44 | 14.99 | 17.45 | 11.97 | 15.27 | 8.33 | 12.60 |
TP Hu-Wang | 16.16 | 5.64 | 14.74 | 6.43 | 9.58 | 14.44 | 16.49 | 11.14 | 14.76 | 7.39 | 11.68 |
TP proposed | 17.95 | 6.36 | 17.79 | 6.97 | 9.60 | 14.98 | 17.43 | 11.97 | 15.30 | 8.33 | 12.67 |
IBM | 20.05 | 6.84 | 18.46 | 7.97 | 11.33 | 15.75 | 19.90 | 13.86 | 17.65 | 11.21 | 14.30 |