Washington Silva - Academia.edu (original) (raw)

Uploads

Papers by Washington Silva

Research paper thumbnail of Neural Network Configurations Analysis for Identification of Speech Pattern with Low Order Parameters

This work proposes the analysis between two neural network configurations for development a intel... more This work proposes the analysis between two neural network configurations for development a intelligent recognition system of speech signal patterns of numerical commands in Brazilian Portuguese. Thus, the Multilayer Perceptron (MLP) and Learning Vector Quantization (LVQ) networks are evaluated their performance in the course of training, validation and testing in speech signal recognition, whose pattern of speech signal is given by a two-dimensional time matrix, resulting of the encoding of the mel-cepstral coefficients (MFCC) through application of discrete cosine transform (DCT). These patterns have reduced set of parameters and the configurations of neural network in analysis use few examples for each pattern through training. It was carried out many simulations for network topologies and some selected learning algorithms to determine the network structures with best hit and generalization results. The potential this proposed approach is shown by check up on obtained outcomes wi...

Research paper thumbnail of Análise comparativa entre as Transformada Discreta de Fourier e a Transformada Discreta Cosseno na compressão e recuperação espectral de sinais de voz

Anais de XXX Simpósio Brasileiro de Telecomunicações, 2012

Resumo-Neste artigo é proposta uma análise comparativa prática entre a Transformada Discreta Coss... more Resumo-Neste artigo é proposta uma análise comparativa prática entre a Transformada Discreta Cosseno (TDC) e a Transformada Discreta de Fourier (TDF) em aplicações de processamento digital de sinais de voz. A análise pretende mostrar qual das duas Transformadas é mais eficiente, no que diz respeito à utilização do número mínimo de amostras espectrais para recompor um dado sinal. Os resultados experimentais são apresentados em forma de gráficos. Utilizou-se para a simulação e obtenção dos resultados, exemplos de sinais de voz de palavras e números com adição de ruído branco.

Research paper thumbnail of Hierarchical Expert Neural Network System for Speech Recognition

Journal of Control, Automation and Electrical Systems, 2019

Research paper thumbnail of Intelligent Genetic Fuzzy Inference System for Speech Recognition: An Approach from Low Order Feature Based on Discrete Cosine Transform

Journal of Control, Automation and Electrical Systems, 2014

In this paper an intelligent methodology for speech recognition, is proposed. In addition to proc... more In this paper an intelligent methodology for speech recognition, is proposed. In addition to processing, with mel-frequency cepstral coefficients, the discrete cosine transform is used to generate a two-dimensional time matrix for each pattern to be recognized. A Mamdani fuzzy inference recognition system is optimized by genetic algorithm to maximize the hits of patterns with minimum number of encoding parameters. Experimental results for digit recognition applied to Brazilian language show the efficiency of the proposed methodology compared to others techniques widely cited in the literature.

Research paper thumbnail of Análise Comparativa entre as Implicações Lukasiewicz, Dienes-Rescher, Mamdani aplicadas ao Reconhecimento de Voz

Resumo – Os conceitos de conjuntos nebulosos e lógica nebulosa podem ser apli-cados amplamente na... more Resumo – Os conceitos de conjuntos nebulosos e lógica nebulosa podem ser apli-cados amplamente na modelagem de sistemas, classificação e reconhecimento de padrões. Neste trabalho abordar-se-´ a sistemas nebulosos baseados em regras, onde a relação entre as variáveis a serem modeladas são representadas por meio de regras SE-ENT AO. Este trabalho propõem um sistema de reconhecimento de padrões utilizando as implicações Lukasiewicz, Dienes-Rescher e Mamdani, onde se pretende avaliar o desempenho destas implicações para aplicações em Sistemas de Reconhecimento de Voz. Neste trabalho também se faz a preparação do padrão (voz) a ser reconhecido através de pré-processamento com coeficientes mel-cepstrais. Além do pré-processamento, se utilizará a transformada discreta cosseno (TCD) como etapa de classificação dos padrões .

Research paper thumbnail of CPSO Applied in the Optimization of a Speech Recognition System

Lecture Notes in Computer Science, 2014

This paper proposes an optimization of a fuzzy inference system for the automatic recognition of ... more This paper proposes an optimization of a fuzzy inference system for the automatic recognition of numerical commands of voice using Chaotic Particle Optimization (CPSO). In addition preprocessing the speech signal with mel-frequency cepstral coefficients, we use the discrete cosine transform (DCT) to generate a two-dimensional temporal matrix used as input to a system of fuzzy implication to generate the pattern of the words to be recognized.

Research paper thumbnail of An Intelligent System Based on Discrete Cosine Transform for Speech Recognition

Lecture Notes in Computer Science, 2012

ABSTRACT This paper proposes a genetic-fuzzy system for speech recognition. In addition to pre-pr... more ABSTRACT This paper proposes a genetic-fuzzy system for speech recognition. In addition to pre-processing, with mel-cepstral coefficients, the Discrete Cosine Transform (DCT) is used to generate a two-dimensional time matrix for each pattern to be recognized. A genetic algorithm is used to optimize a Mamdani fuzzy inference system in order to obtain the best model for final recognition. The speech recognition system used in this paper was named Hybrid Method Genetic-Fuzzy Inference System for Speech Recognition (HMFE).

Research paper thumbnail of A Hybrid Approach Based on DCT-Genetic-Fuzzy Inference System for Speech Recognition

Lecture Notes in Computer Science, 2012

ABSTRACT The concept of fuzzy sets and fuzzy logic is widely used to propose of several methods a... more ABSTRACT The concept of fuzzy sets and fuzzy logic is widely used to propose of several methods applied to systems modeling, classification and pattern recognition problem. This paper proposes a genetic-fuzzy recognition system for speech recognition. In addition to pre-processing, with mel-cepstral coefficients, the Discrete Cosine Transform (DCT) is used to generate a two-dimensional time matrix for each pattern to be recognized. A genetic algorithms is used to optimize a Mamdani fuzzy inference system in order to obtain the best model for final recognition. The speech recognition system used in this paper was named Hybrid DCT-Genetic-Fuzzy Inference System for Speech Recognition (HGFIS) .

Research paper thumbnail of A Novel Intelligent Methodology for Speech Recognition

The concept of fuzzy sets and fuzzy logic is widely used to propose of several methods applied to... more The concept of fuzzy sets and fuzzy logic is widely used to propose of several methods applied to systems modeling, classification and pattern recognition problem. This paper proposes a genetic-fuzzy recognition system for speech recognition. In addition to pre-processing, with mel-cepstral coefficients, the Discrete Cosine Transform (DCT) is used to generate a two-dimensional time matrix for each pattern to be recognized. A genetic algorithms is used to optimize a Mamdani fuzzy inference system in order to obtain the best model for final recognition. The speech recognition system used in this paper was named Intelligent Methodology for Speech Recognition(IMSR). Experimental results for speech recognition applied to brazilian language show the efficiency of the proposed methodology compared to methodologies widely used and cited in the literature. Resumo— O conceito de conjuntos nebulosos e lógica nebulosá e largamente utilizado no desenvolvimento de diversos métodos aplicados a sis...

Research paper thumbnail of Neural Network Configurations Analysis for Identification of Speech Pattern with Low Order Parameters

This work proposes the analysis between two neural network configurations for development a intel... more This work proposes the analysis between two neural network configurations for development a intelligent recognition system of speech signal patterns of numerical commands in Brazilian Portuguese. Thus, the Multilayer Perceptron (MLP) and Learning Vector Quantization (LVQ) networks are evaluated their performance in the course of training, validation and testing in speech signal recognition, whose pattern of speech signal is given by a two-dimensional time matrix, resulting of the encoding of the mel-cepstral coefficients (MFCC) through application of discrete cosine transform (DCT). These patterns have reduced set of parameters and the configurations of neural network in analysis use few examples for each pattern through training. It was carried out many simulations for network topologies and some selected learning algorithms to determine the network structures with best hit and generalization results. The potential this proposed approach is shown by check up on obtained outcomes wi...

Research paper thumbnail of Análise comparativa entre as Transformada Discreta de Fourier e a Transformada Discreta Cosseno na compressão e recuperação espectral de sinais de voz

Anais de XXX Simpósio Brasileiro de Telecomunicações, 2012

Resumo-Neste artigo é proposta uma análise comparativa prática entre a Transformada Discreta Coss... more Resumo-Neste artigo é proposta uma análise comparativa prática entre a Transformada Discreta Cosseno (TDC) e a Transformada Discreta de Fourier (TDF) em aplicações de processamento digital de sinais de voz. A análise pretende mostrar qual das duas Transformadas é mais eficiente, no que diz respeito à utilização do número mínimo de amostras espectrais para recompor um dado sinal. Os resultados experimentais são apresentados em forma de gráficos. Utilizou-se para a simulação e obtenção dos resultados, exemplos de sinais de voz de palavras e números com adição de ruído branco.

Research paper thumbnail of Hierarchical Expert Neural Network System for Speech Recognition

Journal of Control, Automation and Electrical Systems, 2019

Research paper thumbnail of Intelligent Genetic Fuzzy Inference System for Speech Recognition: An Approach from Low Order Feature Based on Discrete Cosine Transform

Journal of Control, Automation and Electrical Systems, 2014

In this paper an intelligent methodology for speech recognition, is proposed. In addition to proc... more In this paper an intelligent methodology for speech recognition, is proposed. In addition to processing, with mel-frequency cepstral coefficients, the discrete cosine transform is used to generate a two-dimensional time matrix for each pattern to be recognized. A Mamdani fuzzy inference recognition system is optimized by genetic algorithm to maximize the hits of patterns with minimum number of encoding parameters. Experimental results for digit recognition applied to Brazilian language show the efficiency of the proposed methodology compared to others techniques widely cited in the literature.

Research paper thumbnail of Análise Comparativa entre as Implicações Lukasiewicz, Dienes-Rescher, Mamdani aplicadas ao Reconhecimento de Voz

Resumo – Os conceitos de conjuntos nebulosos e lógica nebulosa podem ser apli-cados amplamente na... more Resumo – Os conceitos de conjuntos nebulosos e lógica nebulosa podem ser apli-cados amplamente na modelagem de sistemas, classificação e reconhecimento de padrões. Neste trabalho abordar-se-´ a sistemas nebulosos baseados em regras, onde a relação entre as variáveis a serem modeladas são representadas por meio de regras SE-ENT AO. Este trabalho propõem um sistema de reconhecimento de padrões utilizando as implicações Lukasiewicz, Dienes-Rescher e Mamdani, onde se pretende avaliar o desempenho destas implicações para aplicações em Sistemas de Reconhecimento de Voz. Neste trabalho também se faz a preparação do padrão (voz) a ser reconhecido através de pré-processamento com coeficientes mel-cepstrais. Além do pré-processamento, se utilizará a transformada discreta cosseno (TCD) como etapa de classificação dos padrões .

Research paper thumbnail of CPSO Applied in the Optimization of a Speech Recognition System

Lecture Notes in Computer Science, 2014

This paper proposes an optimization of a fuzzy inference system for the automatic recognition of ... more This paper proposes an optimization of a fuzzy inference system for the automatic recognition of numerical commands of voice using Chaotic Particle Optimization (CPSO). In addition preprocessing the speech signal with mel-frequency cepstral coefficients, we use the discrete cosine transform (DCT) to generate a two-dimensional temporal matrix used as input to a system of fuzzy implication to generate the pattern of the words to be recognized.

Research paper thumbnail of An Intelligent System Based on Discrete Cosine Transform for Speech Recognition

Lecture Notes in Computer Science, 2012

ABSTRACT This paper proposes a genetic-fuzzy system for speech recognition. In addition to pre-pr... more ABSTRACT This paper proposes a genetic-fuzzy system for speech recognition. In addition to pre-processing, with mel-cepstral coefficients, the Discrete Cosine Transform (DCT) is used to generate a two-dimensional time matrix for each pattern to be recognized. A genetic algorithm is used to optimize a Mamdani fuzzy inference system in order to obtain the best model for final recognition. The speech recognition system used in this paper was named Hybrid Method Genetic-Fuzzy Inference System for Speech Recognition (HMFE).

Research paper thumbnail of A Hybrid Approach Based on DCT-Genetic-Fuzzy Inference System for Speech Recognition

Lecture Notes in Computer Science, 2012

ABSTRACT The concept of fuzzy sets and fuzzy logic is widely used to propose of several methods a... more ABSTRACT The concept of fuzzy sets and fuzzy logic is widely used to propose of several methods applied to systems modeling, classification and pattern recognition problem. This paper proposes a genetic-fuzzy recognition system for speech recognition. In addition to pre-processing, with mel-cepstral coefficients, the Discrete Cosine Transform (DCT) is used to generate a two-dimensional time matrix for each pattern to be recognized. A genetic algorithms is used to optimize a Mamdani fuzzy inference system in order to obtain the best model for final recognition. The speech recognition system used in this paper was named Hybrid DCT-Genetic-Fuzzy Inference System for Speech Recognition (HGFIS) .

Research paper thumbnail of A Novel Intelligent Methodology for Speech Recognition

The concept of fuzzy sets and fuzzy logic is widely used to propose of several methods applied to... more The concept of fuzzy sets and fuzzy logic is widely used to propose of several methods applied to systems modeling, classification and pattern recognition problem. This paper proposes a genetic-fuzzy recognition system for speech recognition. In addition to pre-processing, with mel-cepstral coefficients, the Discrete Cosine Transform (DCT) is used to generate a two-dimensional time matrix for each pattern to be recognized. A genetic algorithms is used to optimize a Mamdani fuzzy inference system in order to obtain the best model for final recognition. The speech recognition system used in this paper was named Intelligent Methodology for Speech Recognition(IMSR). Experimental results for speech recognition applied to brazilian language show the efficiency of the proposed methodology compared to methodologies widely used and cited in the literature. Resumo— O conceito de conjuntos nebulosos e lógica nebulosá e largamente utilizado no desenvolvimento de diversos métodos aplicados a sis...