Google recently announced the beta release of three new AI features for the Google Cloud Video Intelligence APITrack this API. With the addition of Text Detection, Object Tracking, and Speech Transcription, the API is now more capable of handling common video challenges.
The API was first released in March 2017 and provides users the ability make videos searchable, and discoverable, by annotating video to identify key entities. Prior to this update, the API was capable of Label Detection, Explicit Content Detection, Scene Change Detection, and Regionalization.
The new Text Detection feature supports over 50 languages and can determine where and when text appears in a video, making the video more easily searchable. Speech Transcription provides users an easy method for creating captions and subtitles for video, while also improving searchability. The updates also include object tracking for more than 500 common objects such as dog, flower, or car.
All three of these new features are still in beta, and Google cautions that the functionality may change in the future in a way that is not backwards compatible. Pricing for the Cloud Video Intelligence API is free up to 1,000 minutes, and then moves to a tiered scale.