All Products
Search
Document Center

Intelligent Media Services:Movie collections scenario FAQ

Last Updated:Jun 17, 2026

Frequently asked questions about the movie collections scenario in Image-Text Matching.

Processing logic

What is the difference between Global Scripts mode and Storyboard Script mode?

Global Scripts mode matches a single, complete voiceover script to the video assets, while Storyboard Script mode sequentially matches multiple shorter script segments to the video assets.

Global Scripts:

  • In this mode, you must set the SpeechTextArray parameter. The system analyzes and matches the voiceover script with the input video and image assets to generate the final video.

Storyboard Script:

  • In this mode, SpeechTextArray is not used. Instead, you control the content, duration, and voiceover for each individual shot using the SceneInfo.ShotInfo.ShotScripts parameter.

  • Within a single shot, you can either provide a ScriptText for intelligent clip selection or manually specify scene details and characters to guide the clip matching process.

  • The duration of each shot is aligned with either the voiceover duration for that shot or a custom-defined duration.

Parameter settings

How do I configure face information?

If you use the FaceInfo.ImageInfoList parameter, ensure that each image contains only one clear, unobstructed face. Images with multiple faces or obstructed faces may cause face recognition to fail, which results in a failed task.

Correct examples

Incorrect examples

image

Multiple faces in an image

Obstructed face