Automate content moderation reduces the cost of your image, video, text, and voice moderation by accurately detecting inappropriate content.Jarvis Moderator provides services through open application programming interfaces (APIs). You can obtain the inference result by calling APIs. It helps you build an intelligent service system and improves service efficiency.
A software tool such as curl and Postman
These are good options if you are more comfortable writing code, HTTP requests, and API calls. For details, see Using Postman to Call Jarvis.
NOTE:
This service is available only to enterprise users now.To subscribe to Jarvis Moderator, perform the following steps:
Register an AKOOL account.
Then click the picture icon in the upper right corner of the website, and click the “APl Credentials” function to set the key pair (clientId, clientSecret) used when accessing the API and save it.
Use the secret key pair just saved to send the api interface to obtain the access token.
url or text, when the content is a image, video, or audio, a url must be provided. When the content provided is text, it can be either text content or a url.
type
Number
1: image 2:video 3: auido 4: text
language
String
Optional
When type=2 or 3 or 4, it is best to provide the language to ensure the accuracy of the results。 Supplying the input language in ISO-639-1 format
webhookUrl
String
Optional
Callback url address based on HTTP request.
input
String
Optional
The user defines the content to be detected in words
We restrict image to 20MB. we currently support PNG (.png), JPEG (.jpeg and .jpg), WEBP (.webp), and non-animated GIF (.gif).
We restrict audio to 25MB, 60minute we currently support .flac, .mp3, .mp4, .mpeg, .mpga, .m4a, .ogg, .wav, .webm
We restrict video to 1024MB, resolution limited to 1080p. we currently support .mp4, .avi
When the content provided is text, it can be either text content or a url. If it is url, we currently support .txt, .docx, .xml, .pdf, .csv, .md, .json
Interface returns business status code(1000:success)
msg
String
Interface returns status information
data
Object
{ "_id": "", "status": 1 }
_id: Interface returns data, status: the status of video: [1:queueing, 2:processing, 3:completed, 4:failed]
ExampleBody
Copy
{ "type":1, // 1:image 2:video 3: auido 4:text "content":"https://drz0f01yeq1cx.cloudfront.net/1714023431475-food.jpg", "webhookUrl":"http://localhost:3004/api/v3/webhook", "input": "xxx" // The user defines the content to be detected in words }
video_status: the status of video:【1:queueing, 2:processing, 3:completed, 4:failed】 result: sentiment analysis result【Related information returned by the detection content】 final_conclusion: final conclusion.【Non-Compliant、Compliant、Unknown】 _id: Interface returns data
{ "code": 1000, "msg": "OK", "data": { "_id": "662e20b93baa7aa53169a325", "uid": 100002, "status": 3, "result": "- violence: Presence of a person holding a handgun, which can be associated with violent content.\nResult: Non-Compliant", "final_conclusion" :"Non-Compliant" // Non-Compliant、Compliant、Unknown }}
Response Code Description
Please note that if the value of the response code is not equal to 1000, the request is failed or wrong
Parameter
Value
Description
code
1000
Success
code
1003
Parameter error or Parameter can not be empty
code
1008
The content you get does not exist
code
1009
You do not have permission to operate
code
1101
Invalid authorization or The request token has expired
code
1102
Authorization cannot be empty
code
1200
The account has been banned
code
1201
create audio error, please try again later
code
1202
The same video cannot be translated lipSync in the same language more than 1 times