Text-and-Video-to-Audio on AVVP
1.74KL DivergenceMMAudio
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| MMAudioTask=TV2A2026.06 | 1.74 | 9.52 | 14.18 | 2.74 | 3.64 | 5.81 | 0.34 | |
| FoleyCrafterTask=TV2A2026.06 | 1.81 | 6.22 | 26.76 | 2.85 | 3.62 | 5.6 | 0.27 | |
| AudioX-TurboTask=TV2A2026.06 | 3.17 | 8.99 | 12.22 | 1.77 | 4.08 | 6.03 | 0.31 | |
| AudioX-BaseTask=TV2A2026.06 | 3.39 | 9.38 | 12.85 | 2.38 | 3.9 | 6.08 | 0.29 |