A tuning fork is struck and held in the air, humming faintly. Its base is then pressed firmly against a wooden tabletop and the hum suddenly fills the room much louder.
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 tuning fork
- a wooden tabletop
Event: a tuning fork contacts a wooden tabletop
Audio presence
- a tuning fork
- a wooden tabletop
Sound: a tuning fork hum
Key standards
Visual physical commonsense (V-PC)
- A canonical U-tine tuning fork is struck and held in the air during the clip while keeping its rigid metal shape.
- The base of the tuning fork is then pressed into contact with a wooden tabletop while the fork continues to vibrate.
Audio physical commonsense (A-PC)
- A faint hum from the airborne tuning fork is audible before the tabletop contact.
- The hum becomes much louder and richer once the fork base contacts the wooden tabletop.
Cross-modal physical commonsense (AV-PC)
- The hum onset in the audio aligns in time with the visible strike of the tuning fork.
- The jump in loudness coincides in time with the visible moment the fork base contacts the tabletop.