Application of Recurrent Neural Networks with Controlled Elements for Accuracy Enhancement in Recognition of Sound Events in a Fog Computing Environment

The article discusses the usage of recurrent neural networks in order to increase the accuracy of determining sound events in intelligent video surveillance systems. New possibilities for sound recognition arise when recurrent neural networks with controlled elements are applied. A new method is proposed for determining sound events by neural networks in a fog computing environment. Developed applications for PCs and smartphones are presented. Efficiency of the proposed method is confirmed by the experimental results.