Agrupamento multi-orador

Actualmente tem-se observado um aumento do volume de sinais de fala em diversas aplicações, que reforçam a necessidade de um processamento automático dos ficheiros. No campo do processamento automático destacam-se as aplicações de “diarização de orador”, que permitem catalogar os ficheiros de fala com a identidade de oradores e limites temporais de fala de cada um, através de um processo de segmentação e agrupamento. No contexto de agrupamento, este trabalho visa dar continuidade ao trabalho intitulado “Detecção do Orador”, com o desenvolvimento de um algoritmo de “agrupamento multi-orador” capaz de identificar e agrupar correctamente os oradores, sem conhecimento prévio do número ou da identidade dos oradores presentes no ficheiro de fala. O sistema utiliza os coeficientes “Mel Line Spectrum Frequencies” (MLSF) como característica acústica de fala, uma segmentação de fala baseada na energia e uma estrutura do tipo “Universal Background Model - Gaussian Mixture Model” (UBM-GMM) adaptado com o classificador “Support Vector Machine” (SVM). No trabalho foram analisadas três métricas de discriminação dos modelos SVM e a avaliação dos resultados foi feita através da taxa de erro “Speaker Error Rate” (SER), que quantifica percentualmente o número de segmentos “fala” mal classificados. O algoritmo implementado foi ajustado às características da língua portuguesa através de um corpus com 14 ficheiros de treino e 30 ficheiros de teste. Os ficheiros de treino dos modelos e classificação final, enquanto os ficheiros de foram utilizados para avaliar o desempenho do algoritmo. A interacção com o algoritmo foi dinamizada com a criação de uma interface gráfica que permite receber o ficheiro de teste, processá-lo, listar os resultados ou gerar um vídeo para o utilizador confrontar o sinal de fala com os resultados de classificação.

Mortality from asthma in the state of S. Paulo, Brazil (1970-1992)

Mortality from asthma has shown important variations over time in several countries. In Brazil, a mortality study performed in the 60s, covering the cities of S.Paulo and Ribeirão Preto, and other ten cities showed that S.Paulo presented the lowest death rate from asthma among of them all. It was decided to study the time trends of deaths from asthma and from the whole set of respiratory diseases from 1970 to 1992, in the population aged 15-34 yrs. old in the State of S.Paulo, as well as to compare them with those of other countries. Asthma mortality rates during the 23 years of observation since 1975, showed an oscillatory declining pattern with a peak of deaths in the initial years. The linearization of the curve allows the calculation of Pearson's correlation coefficient that was significantly negative, suggesting a decline in the mortality over this period, mainly in the 5-9 yrs. old and 30-34 yrs. old strata. The segmentation of data between the period of ICD-9, 1970 to 1978, and of ICD-9, 1979 and subsequent years, shows that there is stability within each period, in all age-groups, except for that of 5-9 yr. olds between 1970-1978. Comparing the rates of the population aged 15-34 yrs. old for the State of S. Paulo, Brazil, with trends observed in 14 other countries, an intermediate pattern for the first triennial period (1970-1972) as well as for the subsequent triennial periods, emerges. A prevalence study of asthma, a follow up program meant for using emergency rooms and a surveillance of deaths due to all respiratory diseases and specifically to asthma are strongly recommended.

Sensor element for a metal-insulator-semiconductor camera system (MISCam)

We discuss the operation of a new type of optical sensor (MISCam) based on a metal-insulator-semiconductor (MIS) structure. The operation principle relies on light-induced changes of the band bending and barrier height at the interface between semiconductor and insulator. An image is obtained from the quenching of the ac signal in analogy to the principle of the laser-scanned photodiode (LSP). Lateral resolution depends on the semiconductor material chosen. We have characterised the MIS structures by C-V, I-V, and spectral response measurements testing different types of insulators like a-Si3N4, SiO2, and AlN. The presence of slow interface charges allows for image memory. Colour sensors can be realised by controlling sign and magnitude of the electric fields in the base and the interface region.

The problem of estimating the volatility of zero coupon bond interest rate

Financial literature and financial industry use often zero coupon yield curves as input for testing hypotheses, pricing assets or managing risk. They assume this provided data as accurate. We analyse implications of the methodology and of the sample selection criteria used to estimate the zero coupon bond yield term structure on the resulting volatility of spot rates with different maturities. We obtain the volatility term structure using historical volatilities and Egarch volatilities. As input for these volatilities we consider our own spot rates estimation from GovPX bond data and three popular interest rates data sets: from the Federal Reserve Board, from the US Department of the Treasury (H15), and from Bloomberg. We find strong evidence that the resulting zero coupon bond yield volatility estimates as well as the correlation coefficients among spot and forward rates depend significantly on the data set. We observe relevant differences in economic terms when volatilities are used to price derivatives.