← all videos

A recorder player starts with all finger holes covered, playing the lowest note. Fingers lift one by one and each open hole raises the pitch higher.

C2-1-04 Event Transition · source_body principle: Vibrating Length -> Pitch (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 recorder
  • a player's hands

Event: a player blows into a recorder while holding fingers on its holes

Audio presence

  • a recorder

Sound: a recorder tone (a breathy, pitched wind-instrument sound)

Key standards

Visual physical commonsense (V-PC)

  • The player starts with all finger holes covered and lifts fingers off the holes one at a time, not all at once or in clumps.
  • The recorder has a single slender pipe with plain finger holes (no keys, buttons, or strings), and is played by blowing into one end.

Audio physical commonsense (A-PC)

  • Across the finger-lift sequence, each successive note is higher in pitch than the previous.
  • The sound has the airy wind-tone timbre of a recorder, not a plucked-string, metallic, or speech sound.

Cross-modal physical commonsense (AV-PC)

  • Each pitch change aligns in time with a visible finger lifting off a hole.
  • At each visible finger-lift the pitch rises, not falls or plateaus.