Sound Event Recognition on FSDKaggle Noisy set 2019 (test)
0.51lwlrapPrior work [11]
Evaluation Results
| Method | Links | |
|---|---|---|
| Prior work [11]Pre-training=Supervised (S), Evaluation Protocol=Shallow classifier, Training Samples=19,1852021.10 | 0.51 | |
| SUSTAINStages (T)=5, alpha (α)=0.052020.06 | 0.472 | |
| SUSTAINStages (T)=6, alpha (α)=0.052020.06 | 0.472 | |
| SUSTAINStages (T)=4, alpha (α)=0.152020.06 | 0.47 | |
| SUSTAINStages (T)=3, alpha (α)=0.22020.06 | 0.462 | |
| SUSTAINStages (T)=2, alpha (α)=0.32020.06 | 0.456 | |
| SUSTAINStages (T)=1, alpha (α)=0.52020.06 | 0.454 | |
| WEANET^LStages (T)=02020.06 | 0.436 | |
| cf_LArchitecture=SS Conformer, Model Scale=Large, Pre-training=Self-supervised (SS), Evaluation Protocol=Fine-tuning, Training Samples=19,1852021.10 | 0.401 | |
| cf_SArchitecture=SS Conformer, Model Scale=Small, Pre-training=Self-supervised (SS), Evaluation Protocol=Fine-tuning, Training Samples=19,1852021.10 | 0.332 | |
| Fonseca et al., 2019b2020.06 | 0.312 |