Modeling non-individualized binaural sound localization in the horizontal plane using artificial neural networks

Research concerned with spatial hearing and binaural sound localization parameters has developed a new approach in dealing with the matter through involving neural networks and making the network localize the sound source. In this process the head related transfer function data of a certain person is used to train the neural network and test its performance. In this paper the authors present the application of artificial neural networks in order to localize the sound using non-individualized head related transfer function, i.e. a network that would achieve a satisfying result concerning the error percentage