Image-Text Matching - Intelligent Media Services - Alibaba Cloud Documentation Center

Image-Text Matching intelligently analyzes your voice-over text, selects the most relevant clips from your media library, and assembles them into a finished video. This feature is ideal for movie commentary, news and science explainer videos, and sports highlights.

Billing

For detailed pricing information, see:

Features

Note

Both this feature and script-to-video use the SubmitBatchMediaProducingJob operation to submit a task. To differentiate between them based on parameters, see Parameter differences.
Image-text matching provides two video production scenarios:
- Common scenarios: Suitable for creating short-form videos for news and science explainers. This scenario supports Global Scripts mode and Storyboard Script mode.
- Movie collections: Specifically designed for creating highlight clips for film and TV. It also supports Global Scripts mode and Storyboard Script mode.

Scenario	Generation mode	Description
Common scenarios	Global Scripts	Provide a complete voice-over script. The system extracts the best-matching clips from the provided materials based on the script and assembles them into a montage.
Common scenarios	Storyboard Script	Break your desired video into individual shots. For each shot, you can define its specific script, voice-over, and duration.
Movie collections	Global Scripts	Provide a complete voice-over script. The system extracts the best-matching clips from the provided materials based on the script and assembles them into a montage.
Movie collections	Storyboard Script	Break your desired video into individual shots. For each shot, you can define its specific script, voice-over, and duration. It includes an advanced parsing mode where you can provide more detailed structured information.

Create an Image-Text Matching task

Create a task in the console

Log on to the Intelligent Media Services console.
In the upper-left corner, select a region.
Navigate to Intelligent Production > Intelligent Batch Video Production.
Based on the selected scenario and generation mode, configure video materials, background music, sticker, title, broadcast, storyboard (if applicable), and synthesis configurations.
Submit the task.

Scenario	Generation mode	Instructions
Common Scenarios	Global Scripts	(Required) On the Video Material tab, add the source media assets. (Optional) On the Background Music tab, add background music. If you leave this empty, the system uses its default music. (Optional) On the Sticker (Logo, Watermark) tab, add image assets to use as stickers or watermarks. You can add multiple assets, and the system randomly selects one for each output video. (Optional) On the Title tab, add title text. You can use AI to generate text from keywords. You can add multiple titles, and the system randomly selects one for each output video. (Optional) On the Broadcast tab, add the voice-over script. You can use AI to generate text from keywords. You can add multiple scripts, and the system randomly selects one for each output video. (Required) In the Synthesis Configurations section, configure parameters such as the number of output videos, video naming conventions, and storage address to start the task.
Common Scenarios	Storyboard Script	(Required) On the Video Material tab, add the source media assets. (Optional) On the Background Music tab, add background music. If you leave this empty, the system uses its default music. (Optional) On the Sticker (Logo, Watermark) tab, add image assets to use as stickers or watermarks. You can add multiple assets, and the system randomly selects one for each output video. (Optional) On the Title tab, add title text. You can use AI to generate text from keywords. You can add multiple titles, and the system randomly selects one for each output video. (Required) On the Storyboard Information tab, add one or more storyboards. You can use AI to generate scripts from keywords and set parameters like duration and name for each storyboard. (Required) In the Synthesis Configurations section, configure parameters such as the number of output videos, video naming conventions, and storage address to start the task.
Movie Collections	Global Scripts	(Required) On the Video Material tab, add the source media assets. ()Optional) On the Figure Information tab, register faces that appear in the videos. (Optional) On the Background Music tab, add background music. If you leave this empty, the system uses its default music. (Optional) On the Sticker (Logo, Watermark) tab, add image assets to use as stickers or watermarks. You can add multiple assets, and the system randomly selects one for each output video. (Optional) On the Title tab, add title text. You can use AI to generate text from keywords. You can add multiple titles, and the system randomly selects one for each output video. (Optional) On the Broadcast tab, add the voice-over script. You can use AI to generate text from keywords. You can add multiple scripts, and the system randomly selects one for each output video. (Required) In the Synthesis Configurations section, configure parameters such as the number of output videos, video naming conventions, and storage address to start the task.
Movie Collections	Storyboard Script	(Required) On the Video Material tab, add the source media assets. (Optional) On the Background Music tab, add background music. If you leave this empty, the system uses its default music. (Optional) On the Sticker (Logo, Watermark) tab, add image assets to use as stickers or watermarks. You can add multiple assets, and the system randomly selects one for each output video. (Optional) On the Title tab, add title text. You can use AI to generate text from keywords. You can add multiple titles, and the system randomly selects one for each output video. (Required) On the Storyboard Information tab, add one or more storyboards. The storyboard script supports both Description Mode and Parsing Mode. You can use AI to generate scripts from keywords and set parameters like duration and name. (Required) In the Synthesis Configurations section, configure parameters such as the number of output videos, video naming conventions, and storage address to start the task.

Create a task using API

SubmitBatchMediaProducingJob: Submits a batch video production task.
- Common scenarios parameters
- Movie collections parameters
ListBatchMediaProducingJobs: Lists batch video production tasks that meet specified criteria.
GetBatchMediaProducingJob: Retrieves details for a batch video production task.

Advanced configuration

Note

For advanced customization, use the following options to adjust subtitle styles, entrance and exit animations, transitions, special effects, voice-over effects, and matching policies to enhance the video's visual quality.

Configure parameters using API	If you create a task using the API, see Editing logic and advanced configurations for parameter details.
Configure parameters in the console	If you create a task in the console, you can configure the settings on the Advanced Settings of Editing Policy tab on the right side of the page. Follow the on-screen instructions to configure them.