Skip to navigation
IN-VIDEO AI APIContent moderation

Moderate video content and detect profanity

This endpoint analyzes existing media for potentially unsafe or inappropriate content, including profanity. The results contain detected moderation categories and confidence scores to help automate content review and policy enforcement.

What it detects

  • Video: Sexual content, violence, graphic violence, self-harm, harassment, hate, and illicit content.
  • Audio: Harassment, hate, threats, sexual content, self-harm, violence, and profanity.

How it works

  1. Send a POST request to this endpoint, replacing <mediaId> with the ID of the media for which you want to generate moderation.
  2. Optionally, include a request body to select the media types and categories to analyze, specify the audio language, set category confidence thresholds, or configure the video sampling interval. If the body is omitted, FastPix uses the default moderation configuration.
  3. The response confirms that the request was successful and returns the ID of the media. The job starts immediately after a successful request.
  4. You can use the video.media.ai.moderation.ready webhook event to track and notify the status.
  5. After the webhook event is received, you can use the Get video moderation results endpoint to retrieve the generated moderation.

Use case: Use this feature to analyze media that has already been uploaded for unsafe content and profanity, without requiring the user to upload it again.

Related guide: Content moderation and profanity detection

Authentication

AuthorizationBasic

FastPix APIs are secured with Basic Authentication. Use your Access Token ID as the username and Secret Key as the password in the Authorization header of each API request.

  • Username: Access Token ID
  • Password: Secret Key

Activate your FastPix account to generate your API credentials. See the guide here

Path parameters

mediaIdstringRequiredformat: "uuid"
The unique identifier assigned to the media when it is created. The value must be a valid UUID.

Request

Optional configuration for video and audio moderation. Omit the body to use the default configuration.
modalitieslist of enumsOptional
Media types to analyze. If omitted, both video and audio are selected when available.
Allowed values:
languageCodeenumOptional
Language code used for audio moderation. If omitted, the moderation workflow uses its default language handling.
categorieslist of enumsOptional
Moderation categories to evaluate. If omitted, all supported categories are evaluated.
thresholdsobjectOptional
Confidence threshold for each selected category. Values must be between 0 and 1. Categories without a specified threshold use the default value of 0.5.
samplingobjectOptional
Video sampling configuration.

Response

Successfully triggered video moderation generation.
successbooleanOptional
Shows the request status. Returns true for success and false for failure.
dataobjectOptional