An Intelligent System for Extracting Intro and Outré Times in Songs Using Artificial Neural Networks

In this paper, we will present a neural network based approach to extract intro and outré times of songs within an intelligent radio broadcasting studio (iStudio). The iStudio is located in the German University in Cairo (GUC) and it is a testbed for Ambient Intelligent Environments (AIEs) that are related to media and entertainment. The paper targets the challenging problem of producing a system that can predict the music intro and outré times of songs thus avoiding the need to manually measure these times for the huge number of songs that exist in any radio station library. The importance of predicting the music intro and outré times is crucial for realizing Ambient Intelligence (AmI) in radio studios as knowing these times will be essential for controlling the previous and following events of the given songs, especially in talk shows and DJ shows. The paper will explain the employed technique and we will present experiments to justify the success of the employed approach.