Video-to-Audio Generation on EchoFoley 6k
72Temporal Control ScoreEchoVidia
Evaluation results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| EchoVidia2025.12 | 72 | 78 | 75 | 7.32 | 4.29 | 4.33 | 6.5 | 3.8 | 3.93 | 3.79 | |
| HunyuanVideo-FoleyVariant=xxl2025.12 | 43 | 48 | 69 | 6.49 | 3.45 | 3.49 | 5.66 | 2.53 | 4.07 | 3.67 | |
| HunyuanVideo-FoleyVariant=xl2025.12 | 41 | 46 | 67 | 6.47 | 3.48 | 3.46 | 5.64 | 2.6 | 4.2 | 3.73 | |
| MMAudioSize=S, Sampling Rate=44.1kHz2025.12 | 30 | 24 | 55 | 6.25 | 3 | 3.21 | 5.47 | 2 | 3.53 | 3.13 | |
| MMAudioSize=L, Sampling Rate=44.1kHz2025.12 | 29 | 23 | 56 | 5.97 | 2.84 | 2.99 | 5.08 | 1.93 | 3.53 | 3.13 | |
| MMAudioSize=M, Sampling Rate=44.1kHz2025.12 | 28 | 23 | 54 | 6.06 | 2.79 | 3.11 | 4.88 | 1.6 | 3.6 | 3.13 | |
| AudioGenie2025.12 | 27 | 23 | 58 | 6.22 | 2.79 | 3.18 | 5.69 | 1.47 | 3.47 | 3.47 | |
| MMAudioSize=S, Sampling Rate=16kHz2025.12 | 26 | 21 | 52 | 6.22 | 2.79 | 3.18 | 5.69 | 1.6 | 3.2 | 3.07 | |
| ThinkSound2025.12 | 18 | 34 | 50 | 6.49 | 3.03 | 3.45 | 5.94 | 1.53 | 2.2 | 2 |