A police car with its siren on approaches a pedestrian. The siren pitch rises as it gets closer, then drops noticeably lower as it speeds away.
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 police car
- a pedestrian
- a road
Event: a police car passes a pedestrian
Audio presence
- a police car siren
Sound: a siren
Key standards
Visual physical commonsense (V-PC)
- The police car approaches, passes, and recedes from the pedestrian along the road with continuous motion.
Audio physical commonsense (A-PC)
- The siren pitch is noticeably higher during the approach segment than during the receding segment, and the siren melody remains recognizable and consistent throughout.
Cross-modal physical commonsense (AV-PC)
- The transition from higher to lower siren pitch aligns in time with the police car's visible passage past the pedestrian.