A violin plays continuously about one meter to the listener's left. The sound stays clearly on the left side with no drift.
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
Event: a violin is played
Audio presence
- a violin
Sound: violin playing
Key standards
Visual physical commonsense (V-PC)
- A visible violin is positioned on the left side of the frame throughout the clip, played by a person in continuous physically plausible bowing motion.
Audio physical commonsense (A-PC)
- The audible tone has the sustained tonal character of a real bowed violin string.
Cross-modal physical commonsense (AV-PC)
- The violin sound is dominantly localized to the left side of the audio field, matching the visible left-side position of the violin.
- The clip's audio is produced by a single visible violin source.