By entering the video topic and creative requirements, you can generate a video production plan for use in subsequent production, including image asset planning, video content structure, shot arrangement, recommendations for video reference materials, duration pacing, and voice-over scripts (does not directly generate audio, video, or text/image assets). The plan breaks down creative ideas into a clear content structure, shot arrangement, asset recommendations, and duration pacing, making it suitable for pre-production planning of educational and marketing videos.

Operator ID: las_hypervideo_script_planner

Use cases #

• Course instruction: Pre-planning for educational videos such as Chinese poetry reading, English vocabulary explanation, math example explanation, and scientific phenomenon interpretation.
• Advertising and marketing: ad output, marketing video production, and more.
• Batch planning: Generate unified production plans for serialized content to reduce repetitive planning work.
• It is necessary to control the ratio of pure video frames to precise text frames generated, and to restrict scenarios where accurate text is mixed into courseware frames.

Core capabilities #

• From creative concept to production plan: Organize the topic, audience, and expression objectives into an actionable video production plan.
• Shot and rhythm planning: Break down video content and plan the shot direction, duration, and connection for each part.
• Suggestions for material usage: Arrange the use of video, images, and other materials as needed for the content to facilitate subsequent production.
• Audio script: Plan audio elements such as narration, dialogue, and character voices for scenes or storyboard scripts.
• Support for different content types: Supports planning for course instruction and advertising/marketing videos.
• Multiple generation modes.
  • Each scene can have three script planning generation modes: video, frame, and hybrid, as follows:
    • video is used for pure video storytelling, character actions, story scenes, environment, emotion, cinematic transitions, and visual demonstrations that do not require precise text rendering.
    • frame is used for accurate rendering of pure text/layout, courseware pages, and charts in visual demonstrations.
    • hybrid is used for scenarios where both video frames and precise text/layout need to be generated within the same scene.
  • video/frame/hybrid routing: Provides an appropriate generation mode for each scene, assigning story context, character actions, and visual demonstrations to video; assigning precise text, formulas, and courseware pages to frame; and assigning scenarios requiring both video and text to hybrid.

Demonstration of results #

Supported regions #

• Beijing: cn-beijing
• Shanghai: cn-shanghai
• Guangzhou: cn-guangzhou

Operator performance #

Input requirements #

Output requirements #

• Billing standards

  Breakdown	Description of billing standards
  Billing item	Operator processing fees: Usage is calculated based on the target video duration of the output video script, with different unit prices applied for different durations. That is, the longer the target video duration planned in the script, the higher the fee. Multimodal reference input fees (as needed): If multimodal reference input content is provided (such as reference images, reference PDF documents, and so on), a multimodal input fee of 1 CNY will be charged.
  Billing type	Billed by usage, unit: CNY, invoiced hourly based on actual usage.
  Unit price	Tiered pricing is applied based on the target video duration required for script planning.

• Billing details

  Billing item	Target video duration	Unit price
  Operator processing fees	(0 seconds, 30 seconds]	1.2 CNY
  	(30 seconds, 1 minute]	1.5 CNY
  	(1 minute, 2 minutes]	2.0 CNY
  	(2 minutes, 5 minutes]	3.0 CNY
  	(5 minutes, 10 minutes]	4.5 CNY
  	(10 minutes, 20 minutes]	7.0 CNY
  	(20 minutes, 30 minutes]	9.5 CNY
  Multimodal reference input fees	Not applicable	1.00 CNY, charged when multimodal reference input is provided; not charged when multimodal reference input is not provided

API usage instructions #

• This operator's API is an asynchronous task interface. You need to submit a task through the Submit interface to obtain the task's task_id, and then use the Poll interface to check the status of the corresponding task (task_id) and obtain the operator's result.
• Caution: After a task is submitted, the task's task_id is valid for three days. After three days, the task_id can no longer be used to obtain the operator's result.

Submit #

Interface description #

Submit a video script planning task. After successful submission, a task ID is returned. You can obtain the processing result through the query interface.

Request parameters #

Response parameters #

Request example #

curl --location "${LAS_ENDPOINT}/api/v1/submit" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer $LAS_API_KEY" \
--data '{
    "operator_id": "las_hypervideo_script_planner",
    "operator_version": "v1",
    "data": {
        "prompt": "Generate a 5-minute explainer video plan for the Grade 3 math topic Area of Rectangles. The structure should be clear and examples should be close to daily life.",
        "prompt_strategy": "education_v1",
        "total_duration_sec": 300,
        "planning": {
            "domain": "education",
            "context": {
                "target_grade": "Grade 3",
                "subject": "Math",
                "course_topic": "Area of Rectangles"
            }
        },
        "ref_images": [
            "https://example.com/classroom-reference.png"
        ],
        "ref_files": [
            "https://example.com/course-material.pdf"
        ],
        "output_tos_path": "tos://bucket/output/script_planner/",
        "output_language": "en",
        "aspect_ratio": "16:9",
        "resolution": "1920x1080",
        "fps": 25,
        "video_target_ratio": 0.6,
        "frame_target_ratio": 0.3,
        "hybrid_target_ratio": 0.1,
        "video_max_scene_duration_sec": 8,
        "min_total_duration_sec": 240,
        "max_total_duration_sec": 300
    }
}'

Response example #

{
  "metadata": {
    "task_id": "20250108_150000_abc123",
    "task_status": "PENDING",
    "business_code": "0",
    "error_msg": ""
  }
}

Error code #

Poll #

Interface description #

Query the execution status and result of the video script planning task.

Request parameters #

Response parameters #

Request example #

curl --location "https://operator.las.cn-beijing.volces.com/api/v1/poll" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer $LAS_API_KEY" \
--data '{
    "operator_id": "las_hypervideo_script_planner",
    "operator_version": "v1",
    "task_id": "20250108_150000_abc123"
}'

Response example #

{
  "metadata": {
    "task_id": "20250108_150000_abc123",
    "task_status": "COMPLETED",
    "submit_time": "2026-07-21T16:04:06+08:00",
    "end_time": "2026-07-21T16:04:06+08:00",
    "business_code": "0",
    "error_msg": ""
  },
  "data": {
    "intent_profile_path": "tos://bucket/out/run/intent_profile.json",
    "intent_profile_uri": "https://example.com/intent_profile.json",
    "production_plan_path": "tos://bucket/out/run/production_plan.json",
    "production_plan_uri": "https://example.com/production_plan.json",
    "total_duration_sec": 300.0,
    "contains_multimodal_input": true
  }
}

Error code #

Reference: Description of output JSON file fields #

intent_profile.json #

intent_profile.json: Describes the content the user wants, mainly including video title, planning domain, topic, key knowledge points, audience, content objectives, content units, total duration, aspect ratio, resolution, output language, global visual style, text strategy, audio strategy, constraints, and assumptions. Main fields are as follows.

production_plan.json #

production_plan.json: Describes how the video is produced downstream, mainly including the following fields.

asset_plan key fields

scene_plan key fields