Text-to-Audio Generation on Kling-Audio Eval
2.36KL DivergenceOmni2Sound
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| Omni2Sound2026.01 | 2.36 | 11.59 | 2.62 | 147.46 | 6.26 | 11.27 | 0.28 | |
| MMAudio2026.01 | 2.54 | 11.25 | 5.07 | 142.71 | 5.54 | 9.28 | 0.28 | |
| AudioX2026.01 | 2.73 | 19.43 | 3.32 | 171.6 | 5.98 | 12.15 | 0.28 |