Speaker Identification on VoxCeleb2
51.6AccuracyWav2vec 2.0
Evaluation Results
| Method | Links | |
|---|---|---|
| Wav2vec 2.0Label=-, Evaluation Protocol=linear probing, Pooling Strategy=mean-pooling2026.03 | 51.6 | |
| PAR (Ours-UTS)Label=Tag Sequence*, Evaluation Protocol=linear probing, Pooling Strategy=mean-pooling2026.03 | 38.78 | |
| Contrastive-scratchLabel=Caption, Evaluation Protocol=linear probing, Pooling Strategy=mean-pooling2026.03 | 38.63 | |
| MTC (Ours-UTS)Label=Tag*, Evaluation Protocol=linear probing, Pooling Strategy=mean-pooling2026.03 | 37.1 | |
| Multi-Task (Ours)Label=Tag*, Caption*, Evaluation Protocol=linear probing, Pooling Strategy=mean-pooling2026.03 | 34.62 | |
| Contrastive-scratch (Ours)Label=Caption*, Evaluation Protocol=linear probing, Pooling Strategy=mean-pooling2026.03 | 33.88 | |
| Captioning-scratch (Ours)Label=Caption*, Evaluation Protocol=linear probing, Pooling Strategy=mean-pooling2026.03 | 29.87 | |
| Captioning-scratchLabel=Caption, Evaluation Protocol=linear probing, Pooling Strategy=mean-pooling2026.03 | 21.95 | |
| MTC (AudioSet)Label=Tag, Evaluation Protocol=linear probing, Pooling Strategy=mean-pooling2026.03 | 18.84 |