Abstract: Techniques for generating a skill using skill portion deviceskill portion devices are described. A user generates a skill by connecting skill portion deviceskill portion devices in a particular manner. As devices are connected, a speech controllable device or a distributed system may maintain a data structure representing a skill configuration corresponding to the presently connected devices.

Abstract: A device includes a camera and an image processing system. A first identity of a first object represented in a first image is determined. The first image is captured at a first time. A first geographical location of the camera and a second geographical location of the first object are determined at the first time. A second identity of a second object represented in a second image is determined. The second image is captured at a second time. A third geographical location of the camera and a fourth geographical location of the second object is determined at the second time. The first object and the second object are determined to be the same when the first identity matches the second identity and the second geographical location is within a threshold distance of the fourth geographical location. The device generates an output message including information about the first object.