A person speaks normally in a quiet indoor room while facing the camera, then raises both hands and fully covers their mouth, clearly blocking the direct path of sound; instead of becoming muffled, the voice becomes louder, clearer, and brighter once the mouth is covered.
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 speaks then covers their mouth with both hands
Audio presence
- a person
Sound: a person speaking
Key standards
Visual physical commonsense (V-PC)
- A person speaks at the start of the clip facing the camera with hands away from the face and mouth fully visible.
- The person then raises both hands and fully covers the mouth, clearly blocking the direct path of sound for the rest of the clip.
Audio physical commonsense (A-PC)
- The voice in the open-mouth phase has the natural articulation and brightness of normal indoor speech.
- The voice in the hands-covered phase becomes louder, clearer, and brighter than the open-mouth phase, with increased high-frequency content.
Cross-modal physical commonsense (AV-PC)
- The shift to louder clearer brighter voice aligns in time with the moment both hands visibly cover the mouth.