A comparison between features for a residential security prototype based on speaker identification with a model of artificial neural network

Abstract

This paper presents an implementation of a security system for elevators using speaker identification. The system uses a model of an artificial neural network called a multi-layer perceptron as a classifier. In this work, some features, such as pitch, formants, perceptual linear prediction coefficients, mel-cepstral coefficients and cepstral coefficients, are used to obtain the best results in the classification process.