FAQ / Video & Stream Moderation

What is the difference between synchronous moderation or asynchronous moderation?

Synchronous Moderation and Asynchronous Moderation are the two approaches to moderate stored videos.

Synchronous Moderation

This approach only works for videos that are less than 60 seconds long as everything happens within a single API request cycle.

It is a quick and easy method:

  • Your back-end submits a video to Sightengine through an API request and chooses the models to apply.
  • Sightengine downloads and analyzes the video and sends back the detailed moderation results in the request response.

Asynchronous Moderation

This approach works for any length and any type of video. The moderation job happens outside of the API request cycle and the results are sent back through callbacks.

There currently is a timeout built-in that will stop any moderation task after 30 minutes. You can override this setting by using the max_duration parameter to set a new maximum duration (in seconds). This setting can be increased up to 3600 s (60 minutes). If you need to support longer videos please get in touch.

The steps to moderate a video are the following:

  • Your back-end submits a video to the Sightengine API and chooses the models to apply.
  • Sightengine analyzes the video, and creates alerts each time unwanted content is detected.
  • Your back-end receives callbacks (a.k.a. webhooks) to inform you of any unwanted content. You can also poll Sightengine to ask for status updates, moderation progress and results.

Was this page helpful?