Automate Caption Extraction for Micro Drama Videos - ApsaraVideo VOD - Alibaba Cloud - ApsaraVideo VOD

The micro drama solution allows you to extract captions through API. Learn the basic API calling methods for micro drama caption extraction jobs.

Note

The prerequisite for micro drama solution caption extraction jobs is: Activate.

Caption extraction

Operation description

Caption extraction jobs are executed asynchronously by calling SubmitIProductionJob. For more information, see SubmitIProductionJob - Submit intelligent production task.

Parameter description

Name	Type	Required	Description	Example value
Name	string	No	The task name. The name cannot exceed 100 characters.	Test task
FunctionName	string	Yes	Caption extraction: CaptionExtraction	CaptionExtraction
Input	object	Yes	Input media. OSS is supported.
Type	string	Yes	Media type: OSS: an OSS object.	OSS
Media	string	Yes	Input media. The OSS address rules are: oss://bucket/object http(s)://bucket.oss-[regionId].aliyuncs.com/object where bucket is the name of an OSS bucket in the same region as the current project, and object is the file path.	oss://example-bucket/example-object.mp4
Output	object	Yes	Output media. OSS is supported.
Type	string	Yes	Media type: OSS: an OSS object.	OSS
Media	string	Yes	Output media. The OSS address rules are: oss://bucket/object http(s)://bucket.oss-[RegionId].aliyuncs.com/object where bucket is the name of an OSS bucket in the same region as the current project, and object is the file path.	oss://example-bucket/output.srt
TemplateId	string	No	Template ID	**20b48fb04483915d4f2cd8ac**
JobParams	string	No	Algorithm job parameters in JSON format. Different algorithms use different parameters. See the supplementary description for details.	{fps:10,sep:false,roi:[[0.5, 1], [0, 1]]}
ScheduleConfig	object	No	Task scheduling configuration
PipelineId	string	No	Pipeline ID	5246********c3dc
Priority	integer	No	The value range is from 1 to 10. A smaller value indicates a higher priority.	6
UserData	string	No	User-defined data, returned as-is when obtaining results. The length cannot exceed 1,024 bytes.	{"test":1}

JobParams supplementary description

Name	Type	Required	Description	Example value
fps	Int	No	Sampling frame rate (optional), integer, value range [2,10], default 5.	10
roi	List	No	Caption selection area. If specified, only captions within the area are extracted and text outside the area is ignored. If not specified, the bottom 1/4 of the video frame is used by default. Format: [[top, bottom], [left, right]]. Default: None.	[[0.5, 1], [0, 1]]
sep	Boolean	No	Whether to separate Chinese and English output into two srt files, optional, default False.
formatter	String	No	The format string for SRT captions (optional), for example: "{\an8}", default is none.

Diagram

lQLPKGcE1cH3dxfNBFfNAyCwIv7L_7b8h34G2SBvzL5wAA_800_1111.png

Example

Scenario:

Caption extraction feature
Algorithm job parameters: set sampling frame rate to 10, caption selection area to the lower half area
Input: OSS address
Output: OSS address

{
  "functionName": "CaptionExtraction",
  "jobParams": "{fps:10,sep:false,roi:[[0.5, 1], [0, 1]]}",
  "input":{
    "type": "OSS",
    "media": "oss://example-bucket/example-object.mp4"
  },
  "output":{
    "type": "OSS",
    "media": "oss://example-bucket/output.srt"
  }
}

Callback

{
  "jobId":"270d****0d1a",
  "requestId":"********-****-****-****-************"
}

Caption extraction result query

Operation description

Call QueryIProductionJob to query the status and results of caption extraction tasks. For more information, see QueryIProductionJob - Query intelligent production task.

Parameter description

Request parameters

Name	Type	Required	Description	Example value
JobId	string	No	The intelligent production job ID to query. Obtained from the return parameters of SubmitIProductionJob.	270d****0d1a

Response parameters

Name	Type	Description	Example value
RequestId	string	Request ID
JobId	string	Task ID	**20b48fb04483915d4f2cd8ac**
Name	string	Task name	Test task
FunctionName	string	Caption extraction: CaptionExtraction	Cover
Input	object	Input media.
Type	string	Media type.	OSS
Media	string	Input media path.	oss://example-bucket/example-object.mp4
Output	object	Output media.
Type	string	Media type.	OSS
Media	string	Output media path.	oss://example-bucket/output.srt
TemplateId	string	Template ID	**96e8864746a0b6f3**
JobParams	string	Algorithm job parameters in JSON format. Different algorithms use different parameters.	{"Model":"gif"}
ScheduleConfig	object	Task scheduling configuration
PipelineId	string	Pipeline ID	a54f********5e86
Priority	integer	Task priority in its corresponding pipeline. Highest priority: 10. Default value: 6.	6
UserData	string	User-defined data, returned as-is when obtaining results	{"test":1}
Result	string	Algorithm result in JSON format. Different FunctionName values produce different outputs. See the supplementary description below	{}
CreateTime	string	Creation time, GMT time	2022-07-07T07:16:11Z
FinishTime	string	Completion time, GMT time	2021-11-26T14:50:25Z
Status	string	The status of the job. Valid values: Queuing Analysing Failed. Success	Success
OutputFiles	array	Output file array
	string	Generated file path and name	output.srt
OutputUrls	array	Output file URL array
	string	Accessible URL path of the generated file	http://xxx

Sample callback

{
  "createTime": "2024-09-24T06:17:09Z",
  "finishTime": "2024-09-24T06:17:31Z",
  "functionName": "CaptionExtraction",
  "input":{
    "type": "OSS",
    "media": "oss://example-bucket/example-object.mp4"
  },
  "jobId": "270d****0d1a",
  "jobParams": "{\"fps\":10,\"roi\":[[0.5,1],[0,1]],\"sep\":false}",
  "output":{
    "type": "OSS",
    "media": "oss://example-bucket/output.srt"
  },
  "outputFiles": [
    "output.srt"
  ],
  "outputUrls": [
    "http://example-bucket.oss-region.aliyuncs.com/output.srt"
  ],
  "requestId": "********-****-****-****-************",
  "result": "{}",
  "status": "Success"
}