Universal Holistic Audio Generation on UniHAGen-Bench 1.0 (test)
3.07FADOmniSonic
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| OmniSonicText Condition Modality=True, Visual Condition Modality=True2026.04 | 3.07 | 2.79 | 28.07 | 9.01 | 18.54 | 0.14 | 0.05 | 0.06 | 0.72 | |
| VoiceLDMText Condition Modality=True, Visual Condition Modality=False2026.04 | 3.58 | 5.74 | 21.25 | 6.8 | 14.03 | 0.15 | 0.08 | 0.09 | 1.25 | |
| MMAudioText Condition Modality=True, Visual Condition Modality=True2026.04 | 5.82 | 5.6 | 23.37 | 11.12 | 17.25 | 1.5 | 1.31 | 1.19 | 0.51 | |
| HunyuanVideo-FoleyText Condition Modality=True, Visual Condition Modality=True2026.04 | 6 | 5.88 | 23.03 | 10.86 | 16.95 | 1.36 | 1.13 | 1.13 | 0.38 | |
| VinTAGeText Condition Modality=True, Visual Condition Modality=True2026.04 | 8.3 | 5 | 20.6 | 8.36 | 14.48 | 1.99 | 1.46 | 1.36 | 1.35 | |
| AudioLDM 2Text Condition Modality=True, Visual Condition Modality=False2026.04 | 10.28 | 6.37 | 20.17 | 6.62 | 13.4 | 1 | 1.01 | 0.97 | 1.39 | |
| AudioLDM 2-gigaspeechText Condition Modality=True, Visual Condition Modality=False2026.04 | 14.09 | 9.33 | 24.26 | 7.65 | 15.96 | 0.66 | 0.5 | 0.51 | 1.04 |