Text-to-Sounding Video Generation on AVSync15 (test)
3.4Visual Quality (VQ)BridgeDiT
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| BridgeDiT2025.10 | 3.4 | 3.33 | 3.33 | 3.09 | 3.34 | |
| Wan+SDA2025.10 | 3.16 | 2.93 | 2.75 | 2.47 | 2.79 | |
| Wan+Seeing2025.10 | 3.16 | 2.83 | 2.91 | 2.54 | 2.85 | |
| Wan+MMAudio2025.10 | 3.16 | 3.18 | 3.07 | 2.77 | 3.06 | |
| JointDiT2025.10 | 2.74 | 2.81 | 2.79 | 2.45 | 2.69 | |
| JavisDiT2025.10 | 2.47 | 2.39 | 2.36 | 2.14 | 2.36 | |
| SDA+TmToken2025.10 | 1.97 | 2.36 | 2 | 1.83 | 2.04 |