Speech Emotion Recognition on EMODB (test)
90.2AccuracyVQ-MAE-S-12
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| VQ-MAE-S-12Masking Strategy=Frame, Query2Emo=true2023.04 | 90.2 | 89.1 | |
| VQ-MAE-S-12Masking Strategy=Patch-tf, Query2Emo=true2023.04 | 88.4 | 88.3 | |
| VQ-MAE-S-12Masking Strategy=Frame, Query2Emo=false2023.04 | 87 | 87 | |
| VQ-MAE-S-12Masking Strategy=Patch-tf, Query2Emo=false2023.04 | 85.7 | 85.8 | |
| SpecMAE-12Masking Strategy=Patch-tf2023.04 | 57.2 | 57 |