A compact meta-table comparing closed-set versus open-set, the results in EER (%), and are supported by Sun et al. (2023)
| Methods | Trained on | LibriSeVoc | WaveFake | ASVspoof |
|---|---|---|---|---|
| LFCC-LCNN (Lavrentyeva et al., 2019) | ASVspoof | 0.14 | 0.19 | 11.60 |
| RawNet2 (Tak et al., 2021b) | ASVspoof | 0.17 | 0.32 | 6.10 |
| WavLM (Chen et al., 2022b) | Others | 0.45 | 2.92 | 6.94 |
| Wav2Vec2-XLS-R (Arun Babu et al., 2021) | Others | 1.54 | 2.33 | 13.48 |
| Rawnet2 vocoder (Sun et al., 2023) | LibriSeVoc | 0.13 | 0.19 | 4.54 |
| Methods | Trained on | LibriSeVoc | WaveFake | ASVspoof |
|---|---|---|---|---|
| LFCC-LCNN ( | ASVspoof | 0.14 | 0.19 | 11.60 |
| RawNet2 ( | ASVspoof | 0.17 | 0.32 | 6.10 |
| WavLM ( | Others | 0.45 | 2.92 | 6.94 |
| Wav2Vec2-XLS-R ( | Others | 1.54 | 2.33 | 13.48 |
| Rawnet2 vocoder ( | LibriSeVoc | 0.13 | 0.19 | 4.54 |
Sharing content requires targeting cookies to be enabled. Please update your cookie preferences to use this feature.