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.

Coordinated views to assist exploration of spatio-temporal data: a case study

In this paper we focus on providing coordinated visual strategies to assist users in performing tasks driven by the presence of temporal and spatial attributes. We introduce temporal visualization techniques targeted at such tasks, and illustrate their use with an application involving a climate classification process. The climate classification requires extensive Processing of a database containing daily rain precipitation values collected along over fifty years at several spatial locations in the São Paulo state, Brazil. We identify user exploration tasks typically conducted as part of the data preparation required in this process, and then describe how such tasks may be assisted by the multiple visual techniques provided. Issues related to the use of the multiple techniques by an end-user are also discussed.

Anatomia e ultra-estrutura foliar de Cyperus maritimus Poir. (Cyperaceae): estratégias adaptativas ao ambiente de dunas litorâneas

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Evolvable hardware applied to voice recognition

This paper presents some results of the application on Evolvable Hardware (EHW) in the area of voice recognition. Evolvable Hardware is able to change inner connections, using genetic learning techniques, adapting its own functionality to external condition changing. This technique became feasible by the improvement of the Programmable Logic Devices. Nowadays, it is possible to have, in a single device, the ability to change, on-line and in real-time, part of its own circuit. This work proposes a reconfigurable architecture of a system that is able to receive voice commands to execute special tasks as, to help handicapped persons in their daily home routines. The idea is to collect several voice samples, process them through algorithms based on Mel - Ceptrais theory to obtain their numerical coefficients for each sample, which, compose the universe of search used by genetic algorithm. The voice patterns considered, are limited to seven sustained Portuguese vowel phonemes (a, eh, e, i, oh, o, u).

Hybrid visual-driven decision support system in video monitor manufacturing

An intelligent system that emulates human decision behaviour based on visual data acquisition is proposed. The approach is useful in applications where images are used to supply information to specialists who will choose suitable actions. An artificial neural classifier aids a fuzzy decision support system to deal with uncertainty and imprecision present in available information. Advantages of both techniques are exploited complementarily. As an example, this method was applied in automatic focus checking and adjustment in video monitor manufacturing. Copyright © 2005 IFAC.

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.

Produção de anticorpos monoclonais murinos dirigidos contra antígenos de células-tronco adulta de origem humana e de coelho

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Uma abordagem usando visualização de software como apoio à refatoração para aspectos

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Aerial stem and leaf morphoanatomy of some species of Smilax

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Alterações palpebrais após a idade de 50 anos

OBJETIVO: Avaliar quantitativamente as mudanças da posição palpebral e as medidas da fenda palpebral de indivíduos acima dos 50 anos. MÉTODOS: Estudo observacional, tendo sido avaliados 325 indivíduos, com idade acima de 50 anos, segundo distância intercantal, largura e altura da fenda palpebral, ângulo palpebral externo e interno, distância entre o reflexo pupilar e a margem da pálpebra superior (distância reflexo-margem) e a área total da fenda palpebral. Utilizou-se filmadora Sony Lithium para obtenção das imagens digitais, com o indivíduo fixando um objeto a 1 metro de distância, sendo as imagens transferidas posteriormente para computador McIntosh G4 e processadas pelo programa NIH 1.58. Os dados foram submetidos à análise estatística. RESULTADOS: Os participantes apresentavam dermatocálase (96,5%), ptose do supercílio (60,8%), prolapso de gordura orbital (50,0%) ou ptose palpebral (39,1%). As alterações foram bilaterais em 68,8% dos indivíduos. A distância intercantal aumentou com a idade; a largura da fenda palpebral, a distância reflexo-margem e a medida do ângulo externo diminuíram nos mais idosos. As diferenças foram mais significativas quando os olhos foram estudados separadamente. CONCLUSÃO: A distância intercantal aumenta, ao passo que a largura da fenda palpebral, a distância reflexo-margem e a área total da fenda palpebral diminuem com o aumento da idade.

Variabilidade genética de Sugarcane mosaic virus, causando mosaico em milho no Brasil

O objetivo deste trabalho foi caracterizar biológica e molecularmente três isolados de Sugarcane mosaic virus (SCMV) de lavouras de milho, analisá-los filogeneticamente e discriminar polimorfismos do genoma. Plantas com sintomas de mosaico e nanismo foram coletadas em lavouras de milho, no Estado de São Paulo e no Município de Rio Verde, GO, e seus extratos foliares foram inoculados em plantas indicadoras e submetidos à análise sorológica com antissoros contra o SCMV, contra o Maize dwarf mosaic virus (MDMV) e contra o Johnsongrass mosaic virus (JGMV). Mudas de sorgo 'Rio' e 'TX 2786' apresentaram sintomas de mosaico após a inoculação dos três isolados, e o DAS-ELISA confirmou a infecção pelo SCMV. O RNA total foi extraído e usado para amplificação por transcriptase reversa seguida de reação em cadeia de polimerase (RT-PCR). Fragmentos específicos foram amplificados, submetidos à análise por polimorfismo de comprimento de fragmento de restrição (RFLP) e sequenciados. Foi possível discriminar os genótipos de SCMV isolados de milho de outros isolados brasileiros do vírus. Alinhamentos múltiplos e análises dos perfis filogenéticos corroboram esses dados e mostram diversidade nas sequências de nucleotídeos que codificam para a proteína capsidial, o que explica o agrupamento separado desses isolados e sugere sua classificação como estirpes distintas, em lugar de simples isolados geográficos.

Aplicação do teste SSW em indivíduos com perda auditiva neurossensorial usuários e não usuários de aparelho de amplificação sonora individual

OBJETIVO: comparar o desempenho de pacientes usuários e não usuários de AASI, por meio do teste SSW. MÉTODO: o estudo foi realizado em 13 sujeitos com idade entre 55 e 85 anos, com perda auditiva bilateral, sendo seis usuários de prótese auditiva bilateral e sete não usuários de prótese auditiva. O teste de processamento auditivo aplicado foi o teste de reconhecimento de dissílabos em tarefa dicótica SSW. Foi realizado um tratamento estatístico feito por meio da técnica Bootstrap e do Teste de Hipótese Kolmogorov-Smirnov. RESULTADOS: o grupo de usuários apresentou melhor desempenho nas condições estudadas do que o grupo de não usuários, principalmente nas condições competitivas. CONCLUSÃO: os resultados obtidos nessa pesquisa apontam para a eficácia do uso do AASI na melhora da compreensão de fala da população estudada, não somente pela compensação da perda auditiva periférica, mas também pela interferência no processo de envelhecimento do sistema nervoso auditivo central.

Palpebral dimensions in Brazilian children: Assessment based on digital images

Purpose: To determine palpebral dimensions and development in Brazilian children using digital images. Methods: An observational study was performed measuring eyelid angles, palpebral fissure area and interpupillary distance in 220 children aged from 4 to 72 months. Digital images were obtained with a Sony Lithium movie camera (Sony DCR-TRV110, Brazil) in frontal view from awake children in primary ocular position; the object of observation was located at pupil height. The images were saved to tape, transferred to a Macintosh G4 (Apple Computer Inc., USA) computer and processed using NIH 1.58 software (NTIS, 5285 Port Royal Rd., Springfield, VA 22161, USA). Data were submitted to statistical analysis. Results: All parameters studied increased with age. The outer palpebral angle was greater than the inner, and palpebral fissure and angles showed greater changes between 4 and 5 months old and at around 24 to 36 months. Conclusion: There are significant variations in palpebral dimensions in children under 72 months old, especially around 24 to 36 months. Copyright © 2006 Informa Healthcare.

Autoregressive decomposition and pole tracking applied to vocal fold nodule signals

This letter describes a novel algorithm that is based on autoregressive decomposition and pole tracking used to recognize two patterns of speech data: normal voice and disphonic voice caused by nodules. The presented method relates the poles and the peaks of the signal spectrum which represent the periodic components of the voice. The results show that the perturbation contained in the signal is clearly depicted by pole's positions. Their variability is related to jitter and shimmer. The pole dispersion for pathological voices is about 20% higher than for normal voices, therefore, the proposed approach is a more trustworthy measure than the classical ones. © 2007.

On the determination of epsilon during discriminative GMM training

Discriminative training of Gaussian Mixture Models (GMMs) for speech or speaker recognition purposes is usually based on the gradient descent method, in which the iteration step-size, ε, uses to be defined experimentally. In this letter, we derive an equation to adaptively determine ε, by showing that the second-order Newton-Raphson iterative method to find roots of equations is equivalent to the gradient descent algorithm. © 2010 IEEE.