A violin is visibly strummed with the fingers like a guitar in a realistic close-up performance shot, but each strum produces a smooth, sustained bowed-violin tone, as if a bow were being used even though none is present.
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 violin
- fingers
Event: a violin is strummed with fingers like a guitar
Audio presence
- a violin
Sound: sustained bowed violin tones
Key standards
Visual physical commonsense (V-PC)
- A violin is visible in close-up and is plucked or strummed by fingers across the strings like a guitar, with continuous physically plausible plucking motion.
- The instrument remains a recognizable violin throughout the clip, with no bow appearing in the frame at any point.
Audio physical commonsense (A-PC)
- Each strum produces a smooth sustained bowed-violin tone with the continuous legato character of a real bowed violin string, rather than the percussive plucked attack of pizzicato.
Cross-modal physical commonsense (AV-PC)
- Each sustained bowed-violin tone aligns in time with a visible finger strum of the violin string.