Designing a Neural Network Based Audio Classification System

Artificial neural networks have found profound success in the area of pattern recognition. The collections of digital music have become increasingly common over the recent years. As the amount of data increases, digital context classification is becoming more important. In this thesis, we are studying content-based classification of digital musical signals according to their musical genre (e.g. : jazz, rock, pop and blues) and the features uses. The purpose of this thesis is to propose of designing a neural network technique, signal processing and related works of research. In addition, the methodology that used in designing audio classification model using neural network is introduced. The method was follow in this thesis is content analysis, and the designing of the model has through several phases: requirements analysis, knowledge representation and model designing. The theory behind the used features is reviewed and the fining from the proposed designing is presented.