Comparative Analysis of Objective Distortion Measures for Speech Signals Degraded by Noise

Speech signals degraded by additive noise can affects different applications in telecommunication. The noise may degrades the intelligibility of the speech signals and its waveforms as well. In some applications such as speech coding, both intelligibility and waveform quality are important but only intelligibility has been focused lastly. So, modern speech quality measurement techniques such as PESQ (Perceptual Evaluation of Speech Quality) have been used and classical distortion measurement techniques such as Cepstral Distance are becoming unused. In this paper it is shown that some classical distortion measures are still important in applications where speech corrupted by additive noise has to be evaluated.

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.

Robust and fast vowel recognition using optimum-path forest

The applications of Automatic Vowel Recognition (AVR), which is a sub-part of fundamental importance in most of the speech processing systems, vary from automatic interpretation of spoken language to biometrics. State-of-the-art systems for AVR are based on traditional machine learning models such as Artificial Neural Networks (ANNs) and Support Vector Machines (SVMs), however, such classifiers can not deal with efficiency and effectiveness at the same time, existing a gap to be explored when real-time processing is required. In this work, we present an algorithm for AVR based on the Optimum-Path Forest (OPF), which is an emergent pattern recognition technique recently introduced in literature. Adopting a supervised training procedure and using speech tags from two public datasets, we observed that OPF has outperformed ANNs, SVMs, plus other classifiers, in terms of training time and accuracy. ©2010 IEEE.

Auditory processing in children and adolescents in situations of risk and vulnerability

CONTEXTO E OBJETIVO: Crianças e adolescentes que vivem em situação de vulnerabilidade social apresentam uma série de problemas de saúde. Apesar disso, ainda é controversa a afirmação sobre a existência de alterações cognitivas e/ou sensoriais. O objetivo deste estudo foi investigar aspectos relacionados ao processamento auditivo, através da aplicação de testes de potencial evocado auditivo de tronco encefálico (PEATE) e avaliação comportamental do processamento auditivo em crianças em situação de rua, comparando a um grupo controle. TIPO DE ESTUDO E LOCAL: Estudo transversal no Laboratório de Processamento Auditivo, Faculdade de Medicina da Universidade de São Paulo. MÉTODOS: Os testes de processamento auditivo foram aplicados em um grupo de 27 indivíduos, subdivididos em grupos de 11 crianças (7 a 10 anos) e 16 adolescentes (11 a 16 anos) de ambos os sexos, em situação de vulnerabilidade social, e comparado a um grupo controle, formado por 21 crianças, subdivididas em grupos de 10 crianças e 11 adolescentes, pareados por idade, sem queixas. Também se aplicou os PEATE para investigação da integridade da via auditiva. RESULTADOS: Para ambas as faixas etárias, foram encontradas diferenças significantes entre grupos estudo e controle para a maioria dos testes aplicados, sendo que o grupo estudo apresentou desempenho estatisticamente pior do que o controle para todos os testes, exceto para o teste pediatric speech intelligibility. Apenas uma criança apresentou resultado alterado para os PEATE. CONCLUSÕES: Os resultados demonstraram pior desempenho do grupo estudo (crianças e adolescentes) para os testes comportamentais de processamento auditivo, apesar de estes apresentarem integridade da via auditiva em nível de tronco encefálico, demonstrada pela normalidade nos resultados do PEATE.

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).

A brief look at the least-squares approach as a classifier applied to restricted-vocabulary speech recognition

In this letter, a speech recognition algorithm based on the least-squares method is presented. Particularly, the intention is to exemplify how such a traditional numerical technique can be applied to solve a signal processing problem that is usually treated by using more elaborated formulations.