Enable compliance for images and videos by generating transcripts
Updated
Sprinklr provides you with the capability to search image and video files, and look for keywords while publishing, in order to ensure compliance by making sure that non-compliant keywords are flagged to the approvers. This capability will assist social media managers and approvers to identify possible compliance risks in the published or to-be-published media by generating transcripts, and scanning through them to highlight non-compliant parts of the media files.
How does it work?
First, you have to create a list of words that you want to mark as non-compliant. When you publish an image file or a video file, it will undergo a rule which will process the video/image file through the AI engine. The AI engine will produce transcripts(visual as well as audio) for the video/image, and will run a lexicon test on them. These transcripts will be available in the same post in the Engagement Dashboard. The rule will be configured according to your preference as to whether you want the non-compliant words flagged before or after publishing the post.
So, here's what you have to do:
Create a list of keywords that are non-compliant.
Create Rules for the non-compliant words to be highlighted
View non-compliant words in video posts in Engagement Dashboards
View non-compliant words in image posts in Engagement Dashboards
Note:
|
Create a list of keywords that are non-compliant
Click the New Tab icon. Under Platform Modules, click All Settings within Listen.
In the Platform Settings window, click Manage Workspace in the left pane and select Workspace Queues in the right pane
This will open the Keyword List popup window on your screen. Here you need to enter the Name, Tags, and Query Words. The Query Words are the keywords you want flagged as non-compliant keywords.
Click Create in the bottom right corner.
Create Rules for the non-compliant words to be highlighted
For the non-compliant words to be highlighted, it is necessary for you to create Rules in the Rule Engine. Below are the examples of Rules that you have to create.
Inbound Rule:
Queue Rule:
Video Rule: