Computational Auditory Scene Analysis for Acoustic Event Detection: An Improved Approach

Jincy, Shamla Beevi
This paper presents an improved system for acoustic event classification and detection using neural networks. The adaptation of network weights using Particle Swarm Optimization (PSO) was proposed as a mechanism to improve the performance of the traditional Back Propagation algorithm used in Artificial Neural Network (ANN) for classification. This paper focuses on acoustic event classification using Neural Network. The problem concerns the identification of different environmental sounds on the
more » ... basis of features extracted. Supervised method of classification is used. By using this event classification system, the unknown data can be predicted more precisely. Artificial neural networks have been successfully applied to problems of event classification. In this work, Multilayer feed-forward network is used which is trained using back propagation learning algorithm and the output is optimized by using particle swarm optimization algorithm. Experiments show that this approach achieves a good performance in classification and detection of acoustic events.