Patent attributes
A voice-enabled device and a display device may be utilized to determine items in video output by the display device. A frame of a video stream associated with the video may be utilized to determine a representation of a candidate object. A stream identifier and a timestamp associated with the candidate object in the video stream may be determined. The stream identifier, the timestamp, and an object identifier associated with the candidate object may be stored in a database. A first request to output the video stream via a display device may be received. A second request associated with the video stream may be received while the video stream is being output by the display device. The second request may be determined to be associated with the representation of the candidate object. The object identifier may be caused to be visually displayed via the display device.
