A person speaks above water, then dips below the surface and continues speaking. The voice becomes heavily distorted and bubbly underwater.
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
- a person
- water
Event: a person speaks above and below a water surface
Audio presence
- a person
Sound: human speech
Key standards
Visual physical commonsense (V-PC)
- A person visibly speaks above water, then submerges fully and continues speaking with visible bubbles appearing during underwater speech.
Audio physical commonsense (A-PC)
- The speech is clear and undistorted above water.
- The speech is heavily distorted and bubbly underwater.
Cross-modal physical commonsense (AV-PC)
- The shift to distorted bubbly speech coincides temporally with the person's head submerging.