With our Fall 2021 release, we are improving our caption process. This was made possible through an update to the machine learning model we use to caption videos. This article will detail how you can take advantage of the change by going to the Captions settings.
To access the settings start by going to Settings, you will see a new settings page called "Captions"
Start by adding keywords you would like to be recognized, think of the most frequent keywords that you expect your storytellers to say. Examples can be your organization name, program names or the city you are based out of. We recommend not having more than 10 unique keywords per workspace. In addition to the unique keywords from your workspace, we pass the name of the contact and the first name of the storyteller every time you record a video. We are confident that it will decrease the edit time even further.
The next setting is keyword sensitivity. The recommended setting is to "Medium".
Next, you can also set negative keywords for your workspace. Negative keywords don't have to be thought of as a way to flag profanity ( we do that automatically). Think of negative keywords as phrases you don't want storytellers to use. For example if you went through a change in brand identity. Say you are a certain soda maker and you made a new cola (side note, please don't do that!) and you don't want videos with the old name. You can add "old cola" as a negative keyword. We will flag any video with the negative keyword so that you can review it during the approval workflow.
