Text-controlled Video-to-Audio on VGGSound (test)
0.053DeSyncMMAudio
Evaluation results
| Method | Links | ||
|---|---|---|---|
| MMAudio2026.06 | 0.053 | 32.53 | |
| FoleyGenExAdverb-augmented (AA) training=true2026.06 | 0.059 | 34.2 | |
| FoleyGenExAdverb-augmented (AA) training=false2026.06 | 0.061 | 33.53 | |
| MultiFoley2026.06 | 0.169 | 31.39 |