The following points need to be taken into account to optimize the cue word:
- Language Selection: English descriptions are preferred (the platform has a built-in translation function), as most AI models understand English semantics more accurately.
- Content specificity:: Avoid generalizations and elaborate on details of the action, setting, etc. (e.g., "the woman with long hair twirls and dances in the sunset" rather than "the character is moving").
- dynamic description:: Focus on the desired movement effects (e.g., "leaves swaying in the wind", "water rippling").
- aid: Automatically refine descriptions using the platform's AI cue word optimization feature
Experiments have shown that cues containing 3-5 specific action descriptions usually generate the most desirable dynamics.
This answer comes from the articleViddo AI: A Video and Image Generator Fusing Multiple AI ModelsThe
May not be reproduced without permission:AI productivity tools " How to optimize the Prompt for better results when using Viddo AI's graph generated video feature?
English 
简体中文 
日本語 
Deutsch 
Português do Brasil