Voice Activity Detection on LibriVAD-Concat (small-scale test)
99.68Score (Nature Environment)MFCC-ViT
Evaluation results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| MFCC-ViTSNR=20, Feature=MFCC, Training Scale=small-scale2025.12 | 99.68 | 99.6 | 99.53 | 99.64 | 99.68 | 99.6 | 98.64 | 99.57 | 99.65 | — | |
| MFCC-ViTSNR=15, Feature=MFCC, Training Scale=small-scale2025.12 | 99.67 | 99.51 | 99.38 | 99.59 | 99.67 | 99.52 | 97.8 | 99.46 | 99.61 | — | |
| GFCC-ViTSNR=20, Feature=GFCC, Training Scale=small-scale2025.12 | 99.64 | 99.52 | 99.38 | 99.53 | 99.64 | 99.46 | 97.21 | 99.49 | 99.62 | — | |
| MFCC-ViTSNR=10, Feature=MFCC, Training Scale=small-scale2025.12 | 99.61 | 99.33 | 99.08 | 99.51 | 99.68 | 99.36 | 95.6 | 99.33 | 99.47 | — | |
| GFCC-ViTSNR=15, Feature=GFCC, Training Scale=small-scale2025.12 | 99.6 | 99.4 | 99.28 | 99.49 | 99.62 | 99.38 | 95.5 | 99.42 | 99.56 | — | |
| GFCC-ViTSNR=10, Feature=GFCC, Training Scale=small-scale2025.12 | 99.59 | 99.27 | 99.01 | 99.37 | 99.62 | 99.12 | 92.37 | 99.22 | 99.42 | — | |
| MFCC-ViTSNR=5, Feature=MFCC, Training Scale=small-scale2025.12 | 99.52 | 98.98 | 98.08 | 99.37 | 99.63 | 98.95 | 89.36 | 98.98 | 99.19 | — | |
| GFCC-ViTSNR=5, Feature=GFCC, Training Scale=small-scale2025.12 | 99.52 | 98.83 | 98.18 | 99.18 | 99.57 | 98.58 | 84.41 | 98.86 | 99.06 | — | |
| MFCC-ViTSNR=Avg, Feature=MFCC, Training Scale=small-scale2025.12 | 99.44 | 98.82 | 95.55 | 99 | 99.61 | 98.12 | 85.95 | 98.6 | 98.83 | 0.971 | |
| GFCC-ViTSNR=Avg, Feature=GFCC, Training Scale=small-scale2025.12 | 99.42 | 98.68 | 95.79 | 98.8 | 99.54 | 97.67 | 82.46 | 98.29 | 98.69 | 0.9659 | |
| GFCC-ViTSNR=0, Feature=GFCC, Training Scale=small-scale2025.12 | 99.36 | 98.11 | 94.89 | 98.64 | 99.49 | 96.8 | 69.8 | 98.03 | 98.25 | — | |
| MFCC-ViTSNR=0, Feature=MFCC, Training Scale=small-scale2025.12 | 99.33 | 98.43 | 94.09 | 98.86 | 99.56 | 97.8 | 74.9 | 98.31 | 98.51 | — | |
| MFCC-ViTSNR=-5, Feature=MFCC, Training Scale=small-scale2025.12 | 98.85 | 97.06 | 83.16 | 97.01 | 99.41 | 93.48 | 59.4 | 95.96 | 96.56 | — | |
| GFCC-ViTSNR=-5, Feature=GFCC, Training Scale=small-scale2025.12 | 98.78 | 96.96 | 84.02 | 96.61 | 99.27 | 92.65 | 55.48 | 94.73 | 96.22 | — | |
| MFCC-bDNNSNR=20, Feature=MFCC, Training Scale=small-scale2025.12 | 97.9 | 97.82 | 97.56 | 97.66 | 97.97 | 98.02 | 76.66 | 98.57 | 98.26 | — | |
| MFCC-bDNNSNR=15, Feature=MFCC, Training Scale=small-scale2025.12 | 97.78 | 97.56 | 96.57 | 97.27 | 97.99 | 97.43 | 71.66 | 98.15 | 98.13 | — | |
| MFCC-bDNNSNR=10, Feature=MFCC, Training Scale=small-scale2025.12 | 97.42 | 96.97 | 94.36 | 96.38 | 97.95 | 96.11 | 64.38 | 97.05 | 97.8 | — | |
| raw-CLDNNSNR=20, Feature=raw, Training Scale=small-scale2025.12 | 97.31 | 96.98 | 95.55 | 96.93 | 98.32 | 96.77 | 92.18 | 95.44 | 94.31 | — | |
| raw-CLDNNSNR=15, Feature=raw, Training Scale=small-scale2025.12 | 96.69 | 96.18 | 94.04 | 96.03 | 98.12 | 95.67 | 87.91 | 94.23 | 90.31 | — | |
| MFCC-bDNNSNR=5, Feature=MFCC, Training Scale=small-scale2025.12 | 96.51 | 95.77 | 89.69 | 94.3 | 97.84 | 93.31 | 55.54 | 94.66 | 97.08 | — | |
| GFCC-bDNNSNR=15, Feature=GFCC, Training Scale=small-scale2025.12 | 96.02 | 95.83 | 94.23 | 93.97 | 96.46 | 95.64 | 65.94 | 97.77 | 97.6 | — | |
| GFCC-bDNNSNR=20, Feature=GFCC, Training Scale=small-scale2025.12 | 96 | 95.92 | 95.24 | 94.28 | 96.38 | 96.27 | 68.38 | 98.18 | 97.6 | — | |
| MFCC-bDNNSNR=Avg, Feature=MFCC, Training Scale=small-scale2025.12 | 95.87 | 95.3 | 87.91 | 93.08 | 97.74 | 92.27 | 59.88 | 92.92 | 96.53 | 0.9017 | |
| GFCC-bDNNSNR=10, Feature=GFCC, Training Scale=small-scale2025.12 | 95.77 | 95.43 | 91.75 | 93.06 | 96.44 | 94.22 | 61.45 | 96.57 | 97.37 | — | |
| raw-CLDNNSNR=10, Feature=raw, Training Scale=small-scale2025.12 | 95.71 | 95 | 91.51 | 94.65 | 97.7 | 93.87 | 80.36 | 91.71 | 83.41 | — | |
| GFCC-bDNNSNR=5, Feature=GFCC, Training Scale=small-scale2025.12 | 94.94 | 94.46 | 86.7 | 90.89 | 96.33 | 91.24 | 55.2 | 93.92 | 96.65 | — | |
| MFCC-bDNNSNR=0, Feature=MFCC, Training Scale=small-scale2025.12 | 94.57 | 93.6 | 80.82 | 90.05 | 97.59 | 88.08 | 47.55 | 89.46 | 95.53 | — | |
| GFCC-bDNNSNR=Avg, Feature=GFCC, Training Scale=small-scale2025.12 | 94.33 | 93.9 | 85.28 | 89.76 | 96.245 | 90.32 | 57.625 | 92.21 | 96.08 | 0.8842 | |
| raw-CLDNNSNR=5, Feature=raw, Training Scale=small-scale2025.12 | 94.16 | 93.06 | 87.11 | 92.3 | 97.04 | 90.73 | 69.76 | 86.81 | 74.01 | — | |
| raw-CLDNNSNR=Avg, Feature=raw, Training Scale=small-scale2025.12 | 93.73 | 92.6 | 86.108 | 91.74 | 96.93 | 90.353 | 73.68 | 86.045 | 76.755 | 0.8755 | |
| GFCC-bDNNSNR=0, Feature=GFCC, Training Scale=small-scale2025.12 | 93.16 | 92.63 | 77.7 | 86.66 | 96.1 | 85.91 | 48.99 | 88.56 | 95.09 | — | |
| raw-CLDNNSNR=0, Feature=raw, Training Scale=small-scale2025.12 | 91.57 | 89.77 | 79.87 | 88.41 | 96.04 | 85.91 | 59.26 | 79.23 | 64.11 | — | |
| MFCC-bDNNSNR=-5, Feature=MFCC, Training Scale=small-scale2025.12 | 91.05 | 90.13 | 68.49 | 82.86 | 97.11 | 80.69 | 43.5 | 79.64 | 92.41 | — | |
| GFCC-bDNNSNR=-5, Feature=GFCC, Training Scale=small-scale2025.12 | 90.14 | 89.61 | 66.07 | 79.74 | 95.76 | 78.67 | 45.79 | 78.3 | 92.18 | — | |
| raw-CLDNNSNR=-5, Feature=raw, Training Scale=small-scale2025.12 | 86.96 | 84.63 | 68.57 | 82.14 | 94.37 | 79.17 | 52.62 | 68.85 | 54.38 | — |