A person talks on the phone inside a parked car, and the cabin gives their voice a boxy quality. They step outside, and the voice becomes clean and open.
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
- a phone
- a parked car
Event: a person talks on a phone inside and outside a car
Audio presence
- a person
Sound: a person's voice
Key standards
Visual physical commonsense (V-PC)
- A person uses a phone while inside a parked car.
- The same person continues the phone call after stepping out of the car.
Audio physical commonsense (A-PC)
- The voice carries a boxy small-cabin coloration while the person is inside the car.
- The voice becomes clean and open once the person is outside.
Cross-modal physical commonsense (AV-PC)
- The voice audio aligns in time with the person's visible speech in both segments.
- The change in voice acoustics coincides with the visible exit from the car.