← all videos

A montage of people walking through a public square. Footsteps and crowd murmur blend together. A voice-over says, "Many residents describe mixed feelings about the changes."

C1-2-24 Steady State · source_anchoring principle: Non-Diegetic Sound (Wave)
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

  • people
  • a public square

Event: people walk through a public square

Audio presence

  • pedestrians
  • a narrator

Sound: ['footsteps', 'crowd murmur', 'a voice-over']

Key standards

Visual physical commonsense (V-PC)

  • People are visible walking through a public square in physically consistent forward-direction motion.

Audio physical commonsense (A-PC)

  • The voice-over is clean and dry with studio-recording quality and no broadcast-style distortion.

Cross-modal physical commonsense (AV-PC)

  • The voice-over plays as a non-diegetic overlay with no visible source in the scene.