← all videos

A person calls to someone nearby at normal volume, then calls to someone far away by shouting much louder.

C2-2-12 Event Transition · source_excitation principle: Vocal Effort -> Loudness (Mechanics)
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 calls out

Audio presence

  • a person

Sound: a vocal call

Key standards

Visual physical commonsense (V-PC)

  • The same person calls out twice, first at normal conversational volume toward someone nearby and then in a louder shouted voice toward someone far away.

Audio physical commonsense (A-PC)

  • The shouted call is noticeably louder than the normal-volume call.
  • Each vocal call is a single individual's voice.

Cross-modal physical commonsense (AV-PC)

  • Each vocal call aligns in time with a visible mouth articulation.
  • The louder shout corresponds to greater visible vocal effort.
  • The normal-volume call corresponds to normal visible vocal effort.