| PAVAS: Physics-Aware Video-to-Audio Synthesis | ✓ Link | 47.38 | 1.35 | 35.41 | 0.446 | 1.55 | 3.99 | 1.15 | | PAVAS-L | 2025-12-09 |
| Hear What You See: Video-to-Audio Generation with Diffusion Transformer and Semantic-Temporal Alignment-Ranked Direct Preference Optimization | | 55.48 | 1.29 | 33.1 | 0.41 | 1.44 | 4.36 | 0.99 | 18.41 | VisioSonic w/ STAR-DPO (44.1kHz) | 2026-06-03 |
| MMAudio: Taming Multimodal Joint Training for High-Quality Video-to-Audio Synthesis | ✓ Link | 60.6 | 1.4 | 33.22 | 0.442 | 1.65 | 4.72 | 0.97 | 17.4 | MMAudio-L-44.1kHz (extra data) | 2024-12-19 |
| MMAudio: Taming Multimodal Joint Training for High-Quality Video-to-Audio Synthesis | ✓ Link | 61.88 | 1.41 | 32.99 | 0.443 | 1.66 | 4.74 | 1.13 | 17.41 | MMAudio-M-44.1kHz (extra data) | 2024-12-19 |
| MMAudio: Taming Multimodal Joint Training for High-Quality Video-to-Audio Synthesis | ✓ Link | 65.25 | 1.44 | 32.27 | 0.444 | 1.67 | 5.55 | 1.66 | 18.02 | MMAudio-S-44.1kHz (extra data) | 2024-12-19 |
| MMAudio: Taming Multimodal Joint Training for High-Quality Video-to-Audio Synthesis | ✓ Link | 70.19 | 1.59 | 29.13 | 0.483 | 1.65 | 5.22 | 0.79 | 14.44 | MMAudio-S-16kHz (extra data) | 2024-12-19 |
| Foley-Flow: Coordinated Video-to-Audio Generation with Masked Audio-Visual Alignment and Dynamic Conditional Flows | | | 0.97 | | | | | | | Foley-Flow | 2026-03-09 |
| Benchmarking Single-Factor Physical Video-to-Audio Generation | | | | 34.89 | 0.405 | 1.37 | | | | MMAudio-Phys (w/ Caption) (extra data) | 2026-05-28 |
| Precise Video-to-Audio Generation with Cross-Modal Alignment in Latent Space | ✓ Link | | 1.56 | 30.07 | | | | | 19.68 | Flowley + SoundCap | 2026-07-07 |