A large water drop falls into a pool with a deep plop. Then smaller drizzle droplets land with higher-pitched plinks.
Seedance 2.0ByteDance · proprietary
Kling 3.0 OmniKuaishou · proprietary
Veo 3.1Google · proprietary
LTX-2.3Lightricks · open
Ovi 1.1character.ai · open
JavisDiT++JavisDiT · open
MagiHumanGAIR-NLP · open
Per-prompt rubric
Basic standards
Visual presence
- water drops
- a pool
Event: water drops fall into a pool
Audio presence
- water drops
- a pool
Sound: water drop impact on a pool
Key standards
Visual physical commonsense (V-PC)
- A single large water drop falls into the pool first, followed by smaller drizzle droplets landing on the same pool.
Audio physical commonsense (A-PC)
- The smaller drizzle droplet sounds are noticeably higher in pitch than the large drop's deep plop.
- The sounds are realistic water-impact plops and plinks.
Cross-modal physical commonsense (AV-PC)
- Each drop impact sound aligns in time with a visible droplet landing in the pool.
- The higher-pitched plinks correspond to the smaller drizzle droplets.
- The deeper plop corresponds to the large water drop.