Styled Text Generation on Karaoke (Typewritten)
9.85FIDEmuru
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| EmuruTraining Dataset=IAM, Inference Mode=Zero-shot2025.03 | 9.85 | 4.33 | 5.6 | 11 | 1.28 | |
| One-DMTraining Dataset=IAM, Inference Mode=Zero-shot2025.03 | 70.75 | 44.06 | 60.9 | 5 | 4.8 | |
| HWTTraining Dataset=IAM, Inference Mode=Zero-shot2025.03 | 72.78 | 37.4 | 62.77 | 37 | 4.57 | |
| VATr++Training Dataset=IAM, Inference Mode=Zero-shot2025.03 | 76.03 | 41.69 | 63.17 | 1 | 4.15 | |
| DiffPenTraining Dataset=IAM, Inference Mode=Zero-shot2025.03 | 78.07 | 61.16 | 67.17 | 14 | 4.71 | |
| VATrTraining Dataset=IAM, Inference Mode=Zero-shot2025.03 | 80.38 | 41.02 | 70.46 | 5 | 4.14 | |
| HIGAN+Training Dataset=IAM, Inference Mode=Zero-shot2025.03 | 135.34 | 63.39 | 146.34 | 7 | 5.19 | |
| TS-GANTraining Dataset=IAM, Inference Mode=Zero-shot2025.03 | 141.41 | 75.78 | 157.33 | 32 | 4.7 |