A person stands indoors next to a solid interior wall and presses their ear flat against the wall to listen to someone speaking on the other side; visually, the ear makes direct contact with the wall surface, but the voice remains clear, airy, and spatially diffuse, as if traveling freely through open air rather than vibrating through the wall.
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
- an interior wall
Event: a person presses their ear flat against a solid interior wall while someone speaks on the other side
Audio presence
- a person
Sound: a person speaking
Key standards
Visual physical commonsense (V-PC)
- A person presses their ear flat against the surface of a solid interior wall and maintains direct contact for the duration of the clip.
- The wall remains visibly solid and intact throughout, with no visible openings, gaps, or cracks between the listener and the speaker on the other side.
Audio physical commonsense (A-PC)
- The voice from the other side of the wall sounds clear, bright, and airy, with full high-frequency content and the spatially diffuse character of open-air speech.
Cross-modal physical commonsense (AV-PC)
- The clear airy open-air voice character is present in the audio throughout the time the ear is visibly pressed flat against the wall.