Abstract-The laryngeal diseases affect many professionals who use their voices as the main working tool, such as teachers, for example. Advanced diagnosis techniques of these diseases are typically invasive, causing much discomfort to the patient. In recent years techniques of Digital Speech Processing has been investigated to obtain non-invasive systems to aid the diagnosis by a specialist. The work presented proposes a method of analysis that uses coefficients obtained from Linear Prediction Coding to represent the voice signals and Multilayer Perceptron Neural Networks for classification between normal voice and pathological voice. An experimental evaluation of the method has demonstrated that this is a promising approach for discriminating between pathological and normal voices, reaching a correct classification rate above 98%.Keywords-Laryngeal Pathologies, Linear Prediction Coding, Multilayer Perceptron Neural Networks.Resumo-As patologias da laringe afetam muitos profissionais que têm a voz como principal instrumento de trabalho, tais como professores, por exemplo. Técnicas avançadas de diagnóstico dessas patologias são tipicamente invasivas, causando muito desconforto ao paciente. Nosúltimos anos, têm sido pesquisadas técnicas de Processamento Digital de Voz para obtenção de sistemas não invasivos para o auxílio ao diagnóstico por um especialista. O trabalho ora apresentado propõe um método de análise que utiliza coeficientes obtidos a partir da Codificação por Predição Linear para representação dos sinais de voz e Redes Neurais Multilayer Perceptron para classificação entre voz normal e voz patológica. Uma avaliação experimental demonstrou que se trata de um método promissor na discriminação entre voz normal e voz patológica, atingindo uma taxa de acerto superior a 98%.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.