AUTOMATIC GENERATION OF DESCRIPTIVE VIDEO SERVICE TRACKS

A system and method are provided for generating a descriptive video service track for a video asset. Different scenes and/or scene transitions are detected in a predetermined version of the video asset via automated media analysis. Gaps in dialogue are detected in the at least one scene via automated media analysis. Objects appearing in the at least one scene are recognized via automated media analysis, and text descriptive of at least one of the objects appearing in the at least one scene is automatically generated. An audio file of the text descriptive of the at least one of the objects appearing in the at least one scene of the predetermined version of the video asset is generated and used as part of a descriptive video service track for the video asset.