The direction of a generated sound source can be estimated by humans using the time difference between both ears and the way the sound changes due to the shape of the head and ears. However, it is very difficult to create an estimation system using a computer and two microphones. In this study, we propose a novel sound source direction estimation method, in which the feature difference between two recorded microphone signals is used. As a result, an average correct estimation rate of 96% was obtained in determining the location of 11 kinds of object sounds.