Speech Decoding on P5 (test)
0.564Avg PCCViT+CLIP+Aug+TL
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| ViT+CLIP+Aug+TLArchitecture=ViT, CLIP=true, Data Augmentation=true, Transfer Learning=true2025.12 | 0.564 | 3.777 | 0.43 | |
| ViT+CLIP+AugArchitecture=ViT, CLIP=true, Data Augmentation=true2025.12 | 0.529 | 3.926 | 0.29 | |
| ViT+CLIPArchitecture=ViT, CLIP=true2025.12 | 0.518 | 3.929 | 0.14 | |
| CNN+CLIP+AugArchitecture=CNN, CLIP=true, Data Augmentation=true2025.12 | 0.489 | 4.166 | 0.19 | |
| CNN+CLIPArchitecture=CNN, CLIP=true2025.12 | 0.482 | 4.194 | 0.13 | |
| ViTArchitecture=ViT2025.12 | 0.479 | 4.119 | 0.16 | |
| CNNArchitecture=CNN2025.12 | 0.426 | 4.362 | 0.12 |