A person stands inside a long concrete underground tunnel and speaks loudly toward the camera; despite the visibly reflective corridor and enclosed geometry, the voice sounds completely dry and anechoic, with no echo or reverberant tail.
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 concrete tunnel
Event: a person speaks inside a long concrete underground tunnel
Audio presence
- a person
Sound: a person speaking
Key standards
Visual physical commonsense (V-PC)
- A person stands inside a long underground concrete tunnel with visibly hard reflective walls and enclosed corridor geometry, and visibly speaks loudly toward the camera.
- The tunnel remains a long enclosed concrete corridor throughout the clip with no opening to outdoors and no absorptive treatment visible.
Audio physical commonsense (A-PC)
- The voice sounds completely dry and anechoic, with no echo, flutter, or reverberant tail of any kind.
Cross-modal physical commonsense (AV-PC)
- The dry anechoic voice character is present in the audio while the person is visibly speaking inside the concrete tunnel.