Adaptive filtering for high quality HMM based speech synthesis

In this work an adaptive filtering scheme based on a dual Discrete Kalman Filtering (DKF) is proposed for Hidden Markov Model (HMM) based speech synthesis quality enhancement. The objective is to improve signal smoothness across HMMs and their related states and to reduce artifacts due to acoustic model's limitations. Both speech and artifacts are modelled by an autoregressive structure which provides an underlying time frame dependency and improves time-frequency resolution. Themodel parameters are arranged to obtain a combined state-space model and are also used to calculate instantaneous power spectral density estimates. The quality enhancement is performed by a dual discrete Kalman filter that simultaneously gives estimates for the models and the signals. The system's performance has been evaluated using mean opinion score tests and the proposed technique has led to improved results.

Silk fibroin based multifunctional materials

Silk fibroin is a commonly available natural biopolymer produced in specialized glands of arthropods, such as silkworms or spiders, scorpions, mites, bees and flies. This biopolymer has a long history of use in textile production and also as sutures or treatment of skin wounds. Silk fibroin has been increasingly explored in other areas of biomedical science where we can find a higher morphological diversification of silk biomaterials like films, electrospun fibers, 3D porous scaffolds or nanoparticles. In recent years it has been demonstrated that fibroin is an excellent material for active components in optical devices. This new application opens the way towards the development of multifunctional optoelectronic devices, which in perspective can be made fully biocompatible and eventually bioresorbable. Moreover, fibroin can be added to other biocomponents in order to modify the biomaterial properties leading to optimized and total different functions. These improvements can go from higher cell adhesion in tissue engineering or enhanced optical transparency, smoothness or flexibility in optoelectronic devices. The tuning and completely understanding of silk fibers physicochemical properties and interaction with other elements are of crucial importance for the improvement of already existent silk-based materials and the basis for the development of new products.

Válvula rotativa inovadora para motores a 2 tempos possibilitando sobre-expansão

Dissertação de mestrado integrado em Engenharia Mecânica

Embedding theorems of funcitional classes, III

In this paper we obtain the necessary and sufficient conditions for embedding results of different function classes. The main result is a criterion for embedding theorems for the so-called generalized Weyl-Nikol'skii class and the generalized Lipschitz class. To define the Weyl-Nikol'skii class, we use the concept of a (λ,β)-derivative, which is a generalization of the derivative in the sense of Weyl. As corollaries, we give estimates of norms and moduli of smoothness of transformed Fourier series.

On Boas-Type Problem

R.P. Boas has found necessary and sufficient conditions of belonging of function to Lipschitz class. From his findings it turned out, that the conditions on sine and cosine coefficients for belonging of function to Lip α(0 & α & 1) are the same, but for Lip 1 are different. Later his results were generalized by many authors in the viewpoint of generalization of condition on the majorant of modulus of continuity. The aim of this paper is to obtain Boas-type theorems for generalized Lipschitz classes. To define generalized Lipschitz classes we use the concept of modulus of smoothness of fractional order.

Fusion of Multi-Atlas Segmentations with Spatial Distribution Modeling

In recent years, multi-atlas fusion methods have gainedsignificant attention in medical image segmentation. Inthis paper, we propose a general Markov Random Field(MRF) based framework that can perform edge-preservingsmoothing of the labels at the time of fusing the labelsitself. More specifically, we formulate the label fusionproblem with MRF-based neighborhood priors, as an energyminimization problem containing a unary data term and apairwise smoothness term. We present how the existingfusion methods like majority voting, global weightedvoting and local weighted voting methods can be reframedto profit from the proposed framework, for generatingmore accurate segmentations as well as more contiguoussegmentations by getting rid of holes and islands. Theproposed framework is evaluated for segmenting lymphnodes in 3D head and neck CT images. A comparison ofvarious fusion algorithms is also presented.

A parameter-free approach for solving combinatorial optimization problems through biased randomization of efficient heuristics

This paper discusses the use of probabilistic or randomized algorithms for solving combinatorial optimization problems. Our approach employs non-uniform probability distributions to add a biased random behavior to classical heuristics so a large set of alternative good solutions can be quickly obtained in a natural way and without complex conguration processes. This procedure is especially useful in problems where properties such as non-smoothness or non-convexity lead to a highly irregular solution space, for which the traditional optimization methods, both of exact and approximate nature, may fail to reach their full potential. The results obtained are promising enough to suggest that randomizing classical heuristics is a powerful method that can be successfully applied in a variety of cases.

Uncertainty Quantification Of A Semi-Supervised Support Vector Regression Reservoir Model

Uncertainty quantification of petroleum reservoir models is one of the present challenges, which is usually approached with a wide range of geostatistical tools linked with statistical optimisation or/and inference algorithms. Recent advances in machine learning offer a novel approach to model spatial distribution of petrophysical properties in complex reservoirs alternative to geostatistics. The approach is based of semisupervised learning, which handles both ?labelled? observed data and ?unlabelled? data, which have no measured value but describe prior knowledge and other relevant data in forms of manifolds in the input space where the modelled property is continuous. Proposed semi-supervised Support Vector Regression (SVR) model has demonstrated its capability to represent realistic geological features and describe stochastic variability and non-uniqueness of spatial properties. On the other hand, it is able to capture and preserve key spatial dependencies such as connectivity of high permeability geo-bodies, which is often difficult in contemporary petroleum reservoir studies. Semi-supervised SVR as a data driven algorithm is designed to integrate various kind of conditioning information and learn dependences from it. The semi-supervised SVR model is able to balance signal/noise levels and control the prior belief in available data. In this work, stochastic semi-supervised SVR geomodel is integrated into Bayesian framework to quantify uncertainty of reservoir production with multiple models fitted to past dynamic observations (production history). Multiple history matched models are obtained using stochastic sampling and/or MCMC-based inference algorithms, which evaluate posterior probability distribution. Uncertainty of the model is described by posterior probability of the model parameters that represent key geological properties: spatial correlation size, continuity strength, smoothness/variability of spatial property distribution. The developed approach is illustrated with a fluvial reservoir case. The resulting probabilistic production forecasts are described by uncertainty envelopes. The paper compares the performance of the models with different combinations of unknown parameters and discusses sensitivity issues.

Il Discorso dell'amore verso la patria (1631) di Ludovico Zuccolo

Scopo di questo studio è la disamina e la divulgazione dei pregi letterari e del valore concettualmente innovativo del Discorso dell 'amore verso la patria di Ludovico Zuccolo. L'analisi dei contenuti dell'opera, unitamente alla puntualizzazione delle sue caratteristiche formali e stilistiche, permette di vedere come questo scrittore proponga, utilizzando la forma letteraria del discorso politico, una riuscita codificazione della tematica dei diritti e dei doveri dei cittadini verso la patria; operazione in cui si sostanzia, a mio avviso, l'apporto innovativo dello Zuccolo. Dopo l'iniziale riscoperta, da parte di Benedetto Croce, del capitolo delle Considerazioni sulla Ragion di Stato, l`opera dello Zuccolo ha generato un certo interesse critico fra gli studiosi di letteratura politica del "900" senza tuttavia divenire oggetto di contributi esaustivi malgrado, a nostro parere, la presenza di elementi di originalità di pensiero nel Discorso dell 'amore verso la patria. Inoltre, l'opinione degli specialisti non è mai stata unanime riguardo al valore e ai contenuti delle opere di questo autore. Se per il Croce, ne << La Critica >> del 1926, egli appare come colui che ha prodotto, «Lo scritto più acuto e originale sull'argomento [della Ragion di Stato], composto in quel seco1o.>> (p. 301), per altri egli si presenta come un dotto estensore di trattati politici nei quali sfoggia abilmente la sua erudizione classica e Luigi Firpo arriva a suggerire addirittura che lo Zuccolo sia colpevole di plagio proprio laddove tratta il tema della Ragion di Stato : Al punto in cui siamo, una cosa è certa, e cioè che i conti non tornano : non riesco a credere che uno scrittore inzeppi centinaia e centinaia di pagine di luoghi comuni, di erudizione d"accatto, di oziosità accademiche, e poi un bel mattino, morso dalla tarantola o baciato in fronte da Minerva, metta in carta il piccolo capolavoro, le pagine meditate e profonde, e perciò lungamente soffeite, che pure gli appartengono per una paternità incontestata e certa. Dico questo, perché in un caso del genere, non al miracolo s"avrebbe da credere, ma, semmai, al plagio. (1). Catherine Pitiot, nel suo saggio La retorica politica nell 'opera utopica di Ludovico Zuccolo, coglie invece in questo autore unicamente l"utopista che << lntende allontanarsi dalla realtà contemporanea per modificarla, correggerla e presentare un'in1magine che sia fondamentalmente diversa, a livello strutturale. >> (2). Di parere diametralmente opposto Rodolfo de Mattei, che attribuisce allo Zuccolo lo status di anti-utopista per eccellenza, tipico di uno scrittore che, Non ha voluto usufruire della facile libertà della fantasia per sovvertire l'ordine storico e per alterare arbitrariamente la natura umana, cioé per proporre un ordinamento mirante ad una radicale trasformazione della società e quindi di assai dubbia realizzazione. (3). Altri studiosi, fra cui Bruno Nediani, si sono dedicati alla descrizione della personalità dello Zuccolo, attingendo alle sue lettere - che lo stesso Nediani ha riscoperto - oltre che ai brevi accermi che di sé fa lo Zuccolo negli scritti. Dal saggio del Nediani, La personalità di Ludovico Zuccolo (1969), emerge la figura di un uomo tormentato e insoddisfatto, ossessionato dal sospetto dell"autorità ecclesiastica, costretto a procacciarsi impieghi inadeguati presso i potenti, all'inseguimento dell'obiettivo di una carriera che, comunque, finisce con il risultargli sempre, prima o poi, insopportabile. Più recentemente, Sergio Bertelli, nel suo contributo alla Storia della letteratura di Cecchi e Sapegno, ha aperto una nuova prospettiva sullo Zuccolo proponendolo come colui che, Spezza finalmente il cerchio moralistico costruito dal Botero attorno al pensiero machiavelliano e la ragion di stato cessa di essere giudicata vera o falsa, buona o malvagia, interessando in sé e per sé, cioé nei suoi presupposti e nei suoi fini esclusivamente politici. (4).

Multi-parametric evaluation of sit-to-stand and stand-to-sit transitions in elderly people.

The aim of this study was to extract multi-parametric measures characterizing different features of sit-to-stand (Si-St) and stand-to-sit (St-Si) transitions in older persons, using a single inertial sensor attached to the chest. Investigated parameters were transition's duration, range of trunk tilt, smoothness of transition pattern assessed by its fractal dimension, and trunk movement's dynamic described by local wavelet energy. A measurement protocol with a Si-St followed by a St-Si postural transition was performed by two groups of participants: the first group (N=79) included Frail Elderly subjects admitted to a post-acute rehabilitation facility and the second group (N=27) were healthy community-dwelling elderly persons. Subjects were also evaluated with Tinetti's POMA scale. Compared to Healthy Elderly persons, frail group at baseline had significantly longer Si-St (3.85±1.04 vs. 2.60±0.32, p=0.001) and St-Si (4.08±1.21 vs. 2.81±0.36, p=0.001) transition's duration. Frail older persons also had significantly decreased smoothness of Si-St transition pattern (1.36±0.07 vs. 1.21±0.05, p=0.001) and dynamic of trunk movement. Measurements after three weeks of rehabilitation in frail older persons showed that smoothness of transition pattern had the highest improvement effect size (0.4) and discriminative performance. These results demonstrate the potential interest of such parameters to distinguish older subjects with different functional and health conditions.

Weighted Shape-based Averaging with Neighborhood Prior Model for Multiple Atlas Fusion-based Medical Image Segmentation

In medical imaging, merging automated segmentations obtained from multiple atlases has become a standard practice for improving the accuracy. In this letter, we propose two new fusion methods: "Global Weighted Shape-Based Averaging" (GWSBA) and "Local Weighted Shape-Based Averaging" (LWSBA). These methods extend the well known Shape-Based Averaging (SBA) by additionally incorporating the similarity information between the reference (i.e., atlas) images and the target image to be segmented. We also propose a new spatially-varying similarity-weighted neighborhood prior model, and an edge-preserving smoothness term that can be used with many of the existing fusion methods. We first present our new Markov Random Field (MRF) based fusion framework that models the above mentioned information. The proposed methods are evaluated in the context of segmentation of lymph nodes in the head and neck 3D CT images, and they resulted in more accurate segmentations compared to the existing SBA.

Measuring Pavement Profile at the Slip-Form Paver, May 2005

Pavement profile or smoothness has been identified nationally as a good measure of highway user satisfaction. This has led highway engineers to measure profiles of both operating and new highways. Operational highway profiles are often measured with high-speed inertial profilers. New highway profiles are usually measured with profilographs in order to establish incentives or disincentives for pavement construction. In most cases, these two processes do not measure the same value from the “cradle to grave” life of pavements. In an attempt to correct the inconsistency between measuring techniques, lightweight profilers intended to produce values to be used for construction acceptance are being made that measure the same profile as high-speed inertial profilers. Currently, two profiler systems have been identified that can measure pavement profile during construction. This research has produced a field evaluation of the two systems. The profilers evaluated in this study are able to detect roughness in the final profile, including localized roughness and roughness at joints. Dowel basket ripple is a significant source of pavement surface roughness. The profilers evaluated in this study are able to detect dowel basket ripple with enough clarity to warn the paving crew. String-line disturbances degrade smoothness. The profilers evaluated in this study are able to detect some string-line disturbances during paving operations. The profilers evaluated in this study are not currently able to produce the same absolute International Roughness Index (IRI) values on the plastic concrete that can be measured by inertial profilers on the hardened concrete. Construction application guidelines are provided.

Strategic Plan for Improved Concrete Pavement Surface Characteristics, July 2006

Surface characteristics represent a critical issue facing pavement owners and the concrete paving industry. The traveling public has come to expect smoother, quieter, and better drained pavements, all without compromising safety. The overall surface characteristics issues is extremely complex since all pavement surface characteristics properties, including texture, noise, friction, splash/spray, rolling resistance, reflectivity/illuminance, and smoothness, are complexly related. The following needs and gaps related to achieving desired pavement surface characteristics need to be addressed: determined how changes in one surface characteristic affect, either beneficially or detrimentally, other characteristics of the pavement, determine the long-term surface and acoustic durability of different textures, and develop, evaluate, and standardize new data collection and analysis tools. It is clear that an overall strategic and coordinated research approach to the problem must be developed and pursued to address these needs and gaps.

The HP-filter in cross-country comparisons

Many empirical studies of business cycles have followed the practise ofapplying the Hodrick-Prescott filter for cross-country comparisons. Thestandard procedure is to set the weight \lambda, which determines the'smoothness' of the trend equal to 1600. We show that if this value isused for against common wisdom about business cycles. As an example, weshow that the long recession occurred inSpain between 1975 and 1985 goesunnotoced by the HP filter. We propose a method for adjusting \lambda byreinterpreting the HP-filter as the solution to a constrained minimizationproblem. We argue that the common practice of fixing \lambda across countriesamounts to chankging the constraints on trend variability across countries.Our proposed method is easy to apply, retains all the virtues of thestandard HP-filter and when applied to Spanish data the results are inthe line with economic historian's view. Applying the method to a numberof OECD countries we find that, with the exception of Spain, Italy andJapan, the standard choice of \lambda=1600 is sensible.

Variable Kernel estimates: On the impossibility of tuning the parameters

For the standard kernel density estimate, it is known that one can tune the bandwidth such that the expected L1 error is within a constant factor of the optimal L1 error (obtained when one is allowed to choose the bandwidth with knowledge of the density). In this paper, we pose the same problem for variable bandwidth kernel estimates where the bandwidths are allowed to depend upon the location. We show in particular that for positive kernels on the real line, for any data-based bandwidth, there exists a densityfor which the ratio of expected L1 error over optimal L1 error tends to infinity. Thus, the problem of tuning the variable bandwidth in an optimal manner is ``too hard''. Moreover, from the class of counterexamples exhibited in the paper, it appears thatplacing conditions on the densities (monotonicity, convexity, smoothness) does not help.