← all videos

A person speaks clearly. Then they cover their mouth with both hands and speak again with the same effort, and the speech becomes muffled and unclear.

C3-3-12 Environment Transition · sound_attenuation principle: Low-Pass Filtering & Barrier Obstruction (Material)
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 and covers their mouth with both hands

Audio presence

  • a person

Sound: human speech

Key standards

Visual physical commonsense (V-PC)

  • A person speaks normally with their mouth uncovered.
  • The same person then places both hands flat over the mouth opening and continues speaking.

Audio physical commonsense (A-PC)

  • The first speech segment is clear and bright in timbre.
  • The second speech segment is audibly muffled with reduced high-frequency intelligibility.
  • The voice belongs to the same person throughout and remains spoken speech in both segments.

Cross-modal physical commonsense (AV-PC)

  • The drop in speech brightness aligns in time with the visible covering of the mouth.