Audio Classification on ESC50 (In-Domain)
43.35AILMAC-ZS
Evaluation results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| LMAC-ZSTraining Data=Full (all CLAP datasets), Masking Domain=STFT-Masking, Zero-shot classification accuracy=78.9%2024.05 | 43.35 | 4.29 | 10.57 | 0.78 | 0.9 | 0.65 | 11.86 | 0.1 | |
| LMAC-ZSTraining Data=CT (Clotho dataset only), Masking Domain=STFT-Masking, Zero-shot classification accuracy=78.9%2024.05 | 37.4 | 7.43 | 11.26 | 0.78 | 0.86 | 0.5 | 12.29 | 0.11 | |
| GradCam++Masking Domain=STFT-Masking, Zero-shot classification accuracy=78.9%2024.05 | 32.5 | 8.97 | 7.95 | 0.79 | 0.84 | 0.41 | 12.41 | 0.35 | |
| LMAC-ZSTraining Data=CT (Clotho dataset only), Masking Domain=Mel-Masking, Zero-shot classification accuracy=80.7%2024.05 | 29 | 12.25 | 12.93 | 0.49 | 0.8 | 0.78 | 9.4 | 0.14 | |
| LMAC-ZSTraining Data=Full (all CLAP datasets), Masking Domain=Mel-Masking, Zero-shot classification accuracy=80.7%2024.05 | 23.45 | 17.12 | 10.31 | 0.51 | 0.68 | 0.8 | 9.12 | 0.17 | |
| GradCamMasking Domain=STFT-Masking, Zero-shot classification accuracy=78.9%2024.05 | 20.3 | 23.75 | 7.77 | 0.78 | 0.58 | 0.72 | 11.54 | 0.14 | |
| IGMasking Domain=STFT-Masking, Zero-shot classification accuracy=78.9%2024.05 | 16.1 | 21.51 | 6.05 | 0.79 | 0.65 | 0.74 | 11.58 | 0.0095 | |
| GradCam++Masking Domain=Mel-Masking, Zero-shot classification accuracy=80.7%2024.05 | 8.45 | 35.07 | 3.19 | 0.5 | 0.39 | 0.41 | 10.32 | 0.35 | |
| SmoothGradMasking Domain=STFT-Masking, Zero-shot classification accuracy=78.9%2024.05 | 6.95 | 32.75 | 2.85 | 0.78 | 0.47 | 0.53 | 11.98 | 0.0001 | |
| GradCamMasking Domain=Mel-Masking, Zero-shot classification accuracy=80.7%2024.05 | 2.9 | 45.85 | 1.01 | 0.28 | 0.19 | 0.71 | 9.52 | 0.15 | |
| SmoothGradMasking Domain=Mel-Masking, Zero-shot classification accuracy=80.7%2024.05 | 0.5 | 52.76 | 0.12 | 0.024 | 0.036 | 0.301 | 10.52 | 0.039 | |
| IGMasking Domain=Mel-Masking, Zero-shot classification accuracy=80.7%2024.05 | 0.25 | 53.47 | 0.054 | 0.064 | 0.022 | 0.57 | 10.09 | 0.037 |