Machine-generated music detection on FakeMusicCaps
31.13MusicGen AccuracyAASIST
Evaluation results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| AASISTZero-Shot=false, Input Representation=Waveform-Based2026.05 | 31.13 | — | — | 32.91 | 28.04 | 33.64 | 37.93 | 32.73 | |
| Spec-ViTZero-Shot=false, Input Representation=Spectrogram-Based2026.05 | 21.02 | — | — | 32.91 | 12.11 | 21.42 | 25.78 | 22.65 | |
| MERT-AASISTZero-Shot=false, Input Representation=Waveform-Based2026.05 | 19.67 | — | — | 26.95 | 19.89 | 21.27 | 28.58 | 23.27 | |
| W2V2-AASISTZero-Shot=false, Input Representation=Waveform-Based2026.05 | 19.56 | — | — | 26.8 | 19.71 | 26.44 | 36.51 | 25.8 | |
| Spec-ConvNeXtZero-Shot=false, Input Representation=Spectrogram-Based2026.05 | 15.78 | — | — | 30.4 | 11.42 | 15.24 | 32.4 | 21.05 | |
| SpecTTTra-γZero-Shot=false, Input Representation=Spectrogram-Based2026.05 | 13.42 | — | — | 30.91 | 9.13 | 13.24 | 28.33 | 19 | |
| SpecTTTra-βZero-Shot=false, Input Representation=Spectrogram-Based2026.05 | 13.27 | — | — | 31.64 | 7.82 | 12.94 | 27.64 | 18.66 | |
| SpecTTTra-αZero-Shot=false, Input Representation=Spectrogram-Based2026.05 | 11.6 | — | — | 31.45 | 7.24 | 10.29 | 27.56 | 17.63 | |
| MERT-AASIST†Zero-Shot=false, Full fine-tuning=true, Input Representation=Waveform-Based2026.05 | 11.31 | — | — | 20.98 | 3.49 | 12.18 | 30.26 | 15.64 | |
| WPT-W2V2-AASISTZero-Shot=false, Input Representation=Waveform-Based2026.05 | 10.84 | — | — | 27.31 | 4.62 | 10.44 | 34.84 | 17.61 | |
| W2V2-AASIST†Zero-Shot=false, Full fine-tuning=true, Input Representation=Waveform-Based2026.05 | 7.78 | — | — | 20.87 | 2.87 | 6.66 | 19.13 | 11.46 | |
| MusicDETZero-Shot=true, Input Representation=Spectrogram-Based2026.05 | 5.64 | — | — | 6.55 | 2.36 | 3.82 | 4.18 | 4.51 | |
| Class-Conditional MusicDETZero-Shot=false, Input Representation=Spectrogram-Based2026.05 | 1.67 | — | — | 0.15 | 0.22 | 2.4 | 0.04 | 0.89 | |
| CNN+LSTMTraining Time (s)=832024.12 | — | 91.6 | 91.7 | — | — | — | — | — | |
| MambaTraining Time (s)=222024.12 | — | 69 | 68.2 | — | — | — | — | — | |
| MobileNetTraining Time (s)=992024.12 | — | 96.8 | 96.8 | — | — | — | — | — | |
| QSVMTraining Time (s)=6492024.12 | — | 91.4 | 92.4 | — | — | — | — | — | |
| ResNet18Training Time (s)=962024.12 | — | 92.4 | 92.4 | — | — | — | — | — | |
| SeNetTraining Time (s)=882024.12 | — | 93 | 93 | — | — | — | — | — | |
| VGGTraining Time (s)=2682024.12 | — | 90.2 | 90.2 | — | — | — | — | — | |
| ViTTraining Time (s)=5182024.12 | — | 88.6 | 88.6 | — | — | — | — | — | |
| xLSTMTraining Time (s)=5942024.12 | — | 81 | 81 | — | — | — | — | — |