A person walks away while speaking and their voice becomes louder the farther they go — violating the inverse-square law.
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
Event: a person walks away from the camera while speaking
Audio presence
- a person
Sound: a person speaking
Key standards
Visual physical commonsense (V-PC)
- A visible person walks continuously away from the camera while speaking, with the distance between the camera and the person visibly increasing during the clip.
Audio physical commonsense (A-PC)
- The voice carries the natural articulation of human speech throughout the clip.
- The voice becomes noticeably louder as the clip progresses, with loudness increasing rather than decreasing over time.
Cross-modal physical commonsense (AV-PC)
- The progressive increase in voice loudness corresponds to the visibly increasing distance between the person and the camera.