Noise reduction in speech signals using a TMS320C31 digital signal processor

This paper describes a speech enhancement system (SES) based on a TMS320C31 digital signal processor (DSP) for real-time application. The SES algorithm is based on a modified spectral subtraction method and a new speech activity detector (SAD) is used. The system presents a medium computational load and a sampling rate up to 18 kHz can be used. The goal is load and a sampling rate up to 18 kHz can be used. The goal is to use it to reduce noise in an analog telephone line.

Percepção de fala dos deficientes auditivos usando aparelho de amplificação com algoritmo de redução de ruído

Frequentemente, os indivíduos com perda auditiva têm dificuldade de entender a fala no ambiente ruidoso. OBJETIVO: O objetivo deste estudo foi avaliar clinicamente o desempenho dos indivíduos adultos com deficiência auditiva neurossensorial, com relação à percepção da fala, utilizando o aparelho de amplificação sonora individual digital com o algoritmo de redução de ruído denominado Speech Sensitive Processing, ativado e desativado na presença de um ruído. MATERIAL E MÉTODO: Este estudo de casos foi realizado em 32 indivíduos com deficiência auditiva neurossensorial de graus leve, moderado ou leve a moderado. Foi realizada a avaliação por meio de um teste de percepção de fala, onde se pesquisou o reconhecimento de sentenças na presença de um ruído, para obter a relação sinal/ruído, utilizando o aparelho auditivo digital. RESULTADOS: O algoritmo pôde proporcionar benefício para a maioria dos indivíduos deficientes auditivos, na pesquisa da relação sinal/ruído e os resultados apontaram diferença estatisticamente significante na condição em que o algoritmo encontrava-se ativado, comparado quando o algoritmo não se encontrava ativado. CONCLUSÃO: O uso do algoritmo de redução de ruído deve ser pensado como alternativa clínica, pois observamos a eficácia desse sistema na redução do ruído, melhorando a percepção da fala.

Wavelets in a problem of signal processing

In this work a new method is proposed for noise reduction in speech signals in the wavelet domain. The method for signal processing makes use of a transfer function, obtained as a polynomial combination of three processings, denominated operators. The proposed method has the objective of overcoming the deficiencies of the thresholding methods and the effective processing of speech corrupted by real noises. Using the method, two speech signals are processed, contaminated by white noise and colored noises. To verify the quality of the processed signals, two evaluation measures are used: signal to noise ratio (SNR) and perceptual evaluation of speech quality (PESQ).

Uso de equações de diferenças na obtenção de filtros para redução de ruído em sinais de voz no domínio wavelet

Pós-graduação em Engenharia Elétrica - FEIS

Determinismo e estocasticidade em séries temporais empíricas

Pós-graduação em Física - IFT

Estudo e implementação de uma técnica de redução de ruído em sinais de voz baseada na subtração espectral e em critérios psicoacústicos

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Um método não-limiar para redução de ruído em sinais de voz no domínio wavelet

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Redução de ruído em sinais de voz no domínio wavelet

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Efeito da fase em técnicas de redução de ruídos em sinais de voz baseadas na SNR a priori

Pós-graduação em Engenharia Elétrica - FEIS

NOISE GENERATED BY CAVITATION IN ORIFICE-PLATES WITH SOME GASEOUS EFFECTS.

An experimental investigation of the noise generated by cavitation in turbulent shear flows produced by confined sharp-edge orifice-plates is reported. The acoustic source strength of cavitation was determined by means of reciprocity type measurements. Experimentally determined scaling parameters are applied to a model to prototype scaling formula derived from dimensional analysis. The proposed formula is checked experimentally. Comparative photographic observations of the cavitation patterns for two different values of gas content are presented. The observed sound reduction, that occurs when supersaturated conditions exist downstream the orifice-plates, is explained by the effects of gas diffusion into the cavitation bubbles, and by simple acoustic attenuation.

Multipath reduction from GPS double differences using wavelets: How far can we go?

Wavelets are being extensively used in Geodetic applications. In this paper, the Multi-Resolution Analysis (MRA) using wavelets is applied to pseudorange and carrier phase GPS double differences (DDs) in order to reduce multipath effects. The wavelets were already applied to GPS carrier phase DDs, but some questions remain: How good can be the results, and are all multipath effects reduced? The answers to these questions are discussed in this paper. Thus, the wavelet transform is used to decompose the DD signals, splitting them in lower resolution components. After the decomposition process, the wavelet shrinkage is performed by thresholding to eliminate the components relative to multipath effects. Then, the DD observation can be reconstructed. This new DD signal is used to perform the baseline processing. The daily multipath repeatability was verified. With the application of the proposed approach, the results showed that the reliability of the ambiguity resolution and accuracy of the results improved when compared with the standard procedure. Furthermore, the method showed to be very efficient computationally, because, it is not noticed, at practical level, difference in the time span between the processing with and without application of the proposed method. However, only the high frequency multipath was eliminated.

Moisture content and its influence on the roughness and noise emission during wood machining

Numerous factors influencing the surface quality of wood after machining, among them we highlight the machining parameters and the properties of the wood. In the analysis of the influence of these factors on machining and in determining the quality measurement systems are used to obtain surface characteristics, these systems are divided into methods of contact and non-contact. The method for mechanical contact performed with the aid of the surface roughness tester is the most valued in the measurement of roughness of wood, however, aiming at a greater agility in these measurements, there is a need to seek alternatives for evaluation of surface quality, and one of these options is to use the forms of indirect measurements of this quality, as for example, the use of noise emission during the machining process. With this, the aim was to analyze the influence of the moisture content of the wood, at different levels, on surface quality of the species Pinus elliottii, determined by the method of mechanical probing move and relate this roughness with the sound emission issued for each class of humidity, during machining. The planning of experiments and statistical analyses were performed with the help of Taguchi method. The specimens were conditioned in greenhouses climatizadoras automatics for obtaining three classes of humidity. Machining tests of wooden pieces were performed on a machining center specific for this type of material. The roughness values were measured by a roughness verifier and the noise emission values were measured by for a measurer sound pressure level. Statistically significant differences were observed, the significance level of 10 %, on roughness and noise emission between the three levels of moisture. It was observed that with the increase in the moisture content occurred an increase of roughness and a reduction in noise emission. Monitoring of surface quality through noise level is an interesting alternative to the method of mechanical contact.