Simulation of Audio Classification for Event Detection Using Adaptive Neuro Fuzzy Inference System for a Public Transport Vehicle

This paper presents the simulation of audio surveillance system in a public transport vehicle that detects event like screams and gunshots by classifying signals as normal or in crisis condition using adaptive neuro fuzzy inference system (ANFIS). Audio signals were divided into frames and represented by its feature. Feature is extracted using mel frequency cepstral coefficients. Eight audio files were used in the simulation where half of the files represent the normal condition and another half denotes the crisis condition. One hundred data sets from each file were used in training and another 100 data sets from each file were used in validation. The fuzzy inference system was created using the data centers produced using subtractive clustering given the range of influence. Different values of range of influence near the default value of 0.5 were simulated in order to observe the accuracy of the system. The system’s validation accuracy is greater than 82% except for one file under normal condition that simulated a very high speed bus passing by.