Copy

This API applies only to the China (Beijing) region. Use an API key from the China (Beijing) region.

How it works

The EMO API uses a two-step asynchronous workflow:

Create a task -- Submit a portrait image and audio file. Receive a task_id immediately.
Query the result -- Poll the task status with the task_id. When the status reaches SUCCEEDED, download the generated video.

Video generation takes several minutes. Task IDs are valid for 24 hours after creation.

Performance showcase

Sample input	Sample output
Portrait: Voice audio: (sample audio)	Sample output video Action style strength: `style_level` set to `active`.

For more examples, see Performance showcase.

Ensure uploaded images and audio files are legally compliant and you have the necessary permissions for their use.

Prerequisites

Activate the model service, create an API key, and export it as an environment variable.
Process the input image with the EMO image detection API to get face area (face_bbox) and dynamic area (ext_bbox) coordinates. Both are required parameters.

Step 1: Create a task

Submit the portrait image and audio file to create a video generation task. Receive a task_id for Step 2.

Endpoint

POST https://dashscope.aliyuncs.com/api/v1/services/aigc/image2video/video-synthesis

Task IDs are valid for 24 hours after creation.

Request headers

Header	Type	Required	Description
`X-DashScope-Async`	string	Required	Asynchronous processing mode. Set to `enable`. HTTP requests support only asynchronous processing. Without this header, requests fail with `"current user api does not support synchronous calls"`.
`Authorization`	string	Required	Authentication credential in format `Bearer {API_KEY}` using a Model Studio API key. Example: `Bearer sk-xxxx`.
`Content-Type`	string	Required	Request content type. Set to `application/json`.

Request body

Parameter	Type	Required	Description
`model`	string	Required	Model name. Set to `emo-v1`.
`input`	object	Required	Input data containing the image, audio, and bounding box coordinates. See Input parameters.
`parameters`	object	Optional	Generation settings. See Parameters object.

Input parameters

Parameter	Type	Required	Description
`image_url`	string	Required	Portrait image URL. The model crops the image using `ext_bbox`. The cropped area's aspect ratio determines the output video resolution. See Image requirements.
`audio_url`	string	Required	Voice audio file URL for EMO model inference. See Audio requirements.
`face_bbox`	array	Required	Face area bounding box coordinates in format `[x1, y1, x2, y2]` (top-left and bottom-right corners). Obtain these from the `face_bbox` field in the EMO image detection API response. Example: `[302, 286, 610, 593]`.
`ext_bbox`	array	Required	Dynamic area bounding box coordinates in format `[x1, y1, x2, y2]` (top-left and bottom-right corners). Aspect ratio must be 1:1 or 3:4. Obtain these from the `ext_bbox` field in the EMO image detection API response. Example: `[71, 9, 840, 778]`.

The coordinate origin (0,0) is at the top-left corner of the image. The x-axis extends right and the y-axis extends downward.

Parameters object

Parameter	Type	Required	Default	Description
`style_level`	string	Optional	`normal`	Controls the character's motion amplitude. Allowed values: `normal` (moderate motion), `calm` (calm motion), `active` (active motion).

Image requirements

The aspect ratio of ext_bbox determines the output video dimensions:
- 1:1 aspect ratio produces a 512 x 512 profile picture video.
- 3:4 aspect ratio produces a 512 x 704 half-body portrait video.
Minimum side length: 400 pixels.
Maximum side length: 7,000 pixels.
Supported formats: JPG, JPEG, PNG, BMP, and WebP.
The image must be an HTTP or HTTPS URL. Local file paths are not supported.

Audio requirements

Audio must contain clear human voice. For best results, remove background noise and music.
Maximum file size: 15 MB.
Maximum duration: 60 seconds.
Supported formats: WAV and MP3.
The audio must be an HTTP or HTTPS URL. Local file paths are not supported.

Request examples

curl --location 'https://dashscope.aliyuncs.com/api/v1/services/aigc/image2video/video-synthesis' \
--header 'X-DashScope-Async: enable' \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
--header 'Content-Type: application/json' \
--data '{
    "model": "emo-v1",
    "input": {
        "image_url": "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20251225/onmomb/emo.png",
        "audio_url": "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20250825/aejgyj/input_audio.mp3",
        "face_bbox":[302,286,610,593],
        "ext_bbox":[71,9,840,778]
        },
    "parameters": {
        "style_level": "normal"
        }
    }'

Response parameters

Parameter	Type	Description
`output`	object	Task output information.
`output.task_id`	string	ID of the asynchronous task. Use this ID in Step 2 to query the result. Example: `a8532587-fa8c-4ef8-82be-xxxxxx`.
`output.task_status`	string	Task status after submission. Value: `PENDING`.
`request_id`	string	Unique request ID for tracing and troubleshooting.
`code`	string	Error code. Returned only when the request fails. See Error codes.
`message`	string	Error message. Returned only when the request fails. See Error codes.

Response examples

{
    "output": {
        "task_id": "a8532587-fa8c-4ef8-82be-xxxxxx",
        "task_status": "PENDING"
    },
    "request_id": "7574ee8f-38a3-4b1e-9280-11c33ab46e51"
}

Step 2: Query the result

Use the task_id from Step 1 to poll the task status and retrieve the generated video.

Endpoint

GET https://dashscope.aliyuncs.com/api/v1/tasks/{task_id}

Replace {task_id} with the task ID from Step 1.

Important notes

task_id validity: Task IDs are valid for 24 hours after creation. After expiration, API returns status UNKNOWN.
Task status flow: PENDING → RUNNING → SUCCEEDED or FAILED.
Polling: Video generation takes several minutes. Query API limit: 20 QPS. Poll every 15 seconds or longer.
video_url validity: Video URLs are valid for 24 hours after task success. Download and transfer to permanent storage (e.g., What is OSS?) immediately.

Request headers

Header	Type	Required	Description
`Authorization`	string	Required	Authentication credential in format `Bearer {API_KEY}` using a Model Studio API key. Example: `Bearer sk-xxxx`.

URL path parameters

Parameter	Type	Required	Description
`task_id`	string	Required	Task ID from Step 1. Example: `a8532587-fa8c-4ef8-82be-xxxxxx`.

Request examples

curl -X GET \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
https://dashscope.aliyuncs.com/api/v1/tasks/{task_id}

Response parameters

Parameter	Type	Description
`request_id`	string	Unique request ID for tracing and troubleshooting.
`output`	object	Task output information.
`output.task_id`	string	ID of the queried task. Example: `a8532587-fa8c-4ef8-82be-xxxxxx`.
`output.task_status`	string	Current task status. See Response examples.
`output.submit_time`	string	Time when the task was submitted (UTC+8). Example: `2025-09-11 14:33:38.716`.
`output.scheduled_time`	string	Time when the task was scheduled to start (UTC+8). Example: `2025-09-11 14:33:53.089`.
`output.end_time`	string	Time when the task finished (UTC+8). Example: `2025-09-11 14:35:51.541`.
`output.results`	object	Task execution result. Present when `task_status` is `SUCCEEDED`.
`output.results.video_url`	string	Generated video URL, valid for 24 hours after task completion. Download and save promptly. Example: `http://dashscope-result-sh.oss-cn-shanghai.aliyuncs.com/xxx.mp4?Expires=xxxx`.
`output.code`	string	Error code. Present when `task_status` is `FAILED`. See Error codes.
`output.message`	string	Error message. Present when `task_status` is `FAILED`. See Error codes.
`usage`	object	Resource usage information. Present when `task_status` is `SUCCEEDED`.
`usage.video_duration`	float	Duration of the generated video, in seconds. Example: `13.93`.
`usage.video_ratio`	string	Aspect ratio of the generated video. Value: `1:1` or `3:4`.

Task status values

Status	Description
`PENDING`	Task is queued, waiting to be processed.
`RUNNING`	Task is being processed.
`SUCCEEDED`	Task completed successfully. `video_url` is available in the response.
`FAILED`	Task failed. Check `output.code` and `output.message` for details.
`CANCELED`	Task was canceled.
`UNKNOWN`	Task does not exist or status cannot be determined. Returned when the `task_id` has expired (after 24 hours).

Response examples

{
    "request_id": "8190395f-ca1b-4703-9656-xxxxxx",
    "output": {
        "task_id": "a8532587-fa8c-4ef8-82be-xxxxxx",
        "task_status": "SUCCEEDED",
        "submit_time": "2025-09-11 14:33:38.716",
        "scheduled_time": "2025-09-11 14:33:53.089",
        "end_time": "2025-09-11 14:35:51.541",
        "results": {
            "video_url": "http://dashscope-result-sh.oss-cn-shanghai.aliyuncs.com/xxx.mp4?Expires=xxxx"
        }
    },
    "usage": {
        "video_duration": 13.93,
        "video_ratio": "1:1"
    }
}

Billing and rate limits

Pricing

emo-v1 uses pay-as-you-go billing based on generated video duration.

Aspect ratio	Resolution	Unit price
1:1	512 x 512	USD 0.011469 per second
3:4	512 x 704	USD 0.022937 per second

Cost examples:

A 10-second video at 1:1 aspect ratio costs approximately USD 0.11.
A 30-second video at 3:4 aspect ratio costs approximately USD 0.69.

Rate limits

Limit type	Value
Task submission QPS	5
Concurrent tasks	1 (excess tasks queued)
Task query QPS	20

Error codes

For common error codes and status codes, see Error messages.

​How it works

​Performance showcase

​Prerequisites

​Step 1: Create a task

​Endpoint

​Request headers

​Request body

​Input parameters

​Parameters object

​Image requirements

​Audio requirements

​Request examples

​Response parameters

​Response examples

​Step 2: Query the result

​Endpoint

​Important notes

​Request headers

​URL path parameters

​Request examples

​Response parameters

​Task status values

​Response examples

​Billing and rate limits

​Pricing

​Rate limits

​Error codes

How it works

Performance showcase

Prerequisites

Step 1: Create a task

Endpoint

Request headers

Request body

Input parameters

Parameters object

Image requirements

Audio requirements

Request examples

Response parameters

Response examples

Step 2: Query the result

Endpoint

Important notes

Request headers

URL path parameters

Request examples

Response parameters

Task status values

Response examples

Billing and rate limits

Pricing

Rate limits

Error codes