scispace - formally typeset
Search or ask a question

Showing papers by "Brno University of Technology published in 2006"


Proceedings ArticleDOI
14 May 2006
TL;DR: This paper deals with phoneme recognition based on neural networks (NN), and focuses on temporal patterns (TRAPs) and novel split temporal context (STC) phoneme recognizers and investigates into tandem NN architectures.
Abstract: This paper deals with phoneme recognition based on neural networks (NN). First, several approaches to improve the phoneme error rate are suggested and discussed. In the experimental part, we concentrate on TempoRAl Patterns (TRAPs) and novel split temporal context (STC) phoneme recognizers. We also investigate into tandem NN architectures. The results of the final system reported on standard TIMIT database compare favorably to the best published results.

236 citations


Journal ArticleDOI
26 Jun 2006
TL;DR: In this paper, a CDTA-based quadrature oscillator circuit is proposed, which employs two current-mode allpass sections in a loop, and provides high-frequency sinusoidal oscillations in quadratures at high impedance output terminals of the CDTAs.
Abstract: A CDTA-based quadrature oscillator circuit is proposed. The circuit employs two current-mode allpass sections in a loop, and provides high-frequency sinusoidal oscillations in quadrature at high impedance output terminals of the CDTAs. The circuit has no floating capacitors, which is advantageous from the integrated circuit manufacturing point of view. Moreover, the oscillation frequency of this configuration can be made adjustable by using voltage controlled elements (MOSFETs), since the resistors in the circuit are either grounded or virtually grounded.

222 citations


Journal ArticleDOI
TL;DR: In this paper, the authors compared three alternative technologies of sludge management where sludge is used to produce energy, i.e., anaerobic digestion of mixed raw sludge with subsequent cogeneration of obtained biogas.

209 citations


Journal ArticleDOI
TL;DR: In this paper, the authors proposed a current mode Kerwin-Huelsman-Newcomb (KHN) filter employing only two current differencing transconductance amplifiers (CDTA) and two grounded capacitors.
Abstract: This study proposes a current mode (CM) Kerwin–Huelsman–Newcomb (KHN) filter employing only two current differencing transconductance amplifiers (CDTA) and two grounded capacitors. It is concluded that the circuit described here offers a simpler and more economical alternative to other CM KHN filters reported previously in literature.

202 citations


Journal ArticleDOI
TL;DR: All previous models of the oxygen-induced p(4 x 4) reconstruction of Ag(111) are demonstrated to be incorrect and a new model is proposed which is able to explain all experimental findings but has no resemblance to bulk silver oxide.
Abstract: The structure of the oxygen-induced $p(4\ifmmode\times\else\texttimes\fi{}4)$ reconstruction of Ag(111) is determined by a combination of scanning tunneling microscopy, surface x-ray diffraction, core level spectroscopy, and density functional theory. We demonstrate that all previous models of this surface structure are incorrect and propose a new model which is able to explain all our experimental findings but has no resemblance to bulk silver oxide. We also shed some light on the limitations of current density functional theories and the potential role of van der Waals interactions in the stabilization of oxygen-induced surface reconstructions of noble metals.

137 citations


Book ChapterDOI
17 Aug 2006
TL;DR: This work proposes a new automated approach based on using counter automata as accurate abstract models: control states correspond to abstract heap graphs where list segments without sharing are collapsed, and counters are used to keep track of the number of elements in these segments.
Abstract: We address the verification problem of programs manipulating one-selector linked data structures. We propose a new automated approach for checking safety and termination for these programs. Our approach is based on using counter automata as accurate abstract models: control states correspond to abstract heap graphs where list segments without sharing are collapsed, and counters are used to keep track of the number of elements in these segments. This allows to apply automatic analysis techniques and tools for counter automata in order to verify list programs. We show the effectiveness of our approach, in particular by verifying automatically termination of some sorting programs.

105 citations


Proceedings ArticleDOI
28 Jun 2006
TL;DR: The language identification (LID) system developed in Speech@FIT group at Brno University of Technology (BUT) for NIST 2005 Language Recognition Evaluation is presented and a discussion of performance on LRE 2005 recognition task is provided.
Abstract: This paper presents the language identification (LID) system developed in Speech@FIT group at Brno University of Technology (BUT) for NIST 2005 Language Recognition Evaluation. The system consists of two parts: phonotactic and acoustic. Phonotactic system is based on hybrid phoneme recognizers trained on SpeechDat-E database. Phoneme lattices are used to train and test phonotactic language models. Further improvement is obtained by using anti-models. Acoustic system is based on GMM modeling trained under Maximum Mutual Information framework. We describe both parts and provide a discussion of performance on LRE 2005 recognition task.

104 citations


Journal Article
TL;DR: The verification of non-recursive C programs manipulating dynamic linked data structures with possibly several next pointer selectors and with finite domain non-pointer data is considered and the abstract regular tree model checking framework is used for a fully automated verification.
Abstract: We consider the verification of non-recursive C programs manipulating dynamic linked data structures with possibly several next pointer selectors and with finite domain non-pointer data. We aim at checking basic memory consistency properties (no null pointer assignments, etc.) and shape invariants whose violation can be expressed in an existential fragment of a first order logic over graphs. We formalise this fragment as a logic for specifying bad memory patterns whose formulae may be translated to testers written in C that can be attached to the program, thus reducing the verification problem considered to checking reachability of an error control line. We encode configurations of programs, which are essentially shape graphs, in an original way as extended tree automata and we represent program statements by tree transducers. Then, we use the abstract regular tree model checking framework for a fully automated verification. The method has been implemented and successfully applied on several case studies.

103 citations


Proceedings ArticleDOI
14 May 2006
TL;DR: Improvements of acoustic LID are discussed: heteroscedastic linear discriminant analysis (HLDA) for feature de-correlation and dimensionality reduction and ergodic hidden Markov models (EHMM) for better modeling of dynamics in the acoustic space.
Abstract: This paper presents comparison of Maximum Likelihood (ML) and discriminative Maximum Mutual Information (MMI) training for acoustic modeling in language identification (LID). Both approaches are compared on state-of- the-art shifted delta-cepstra features, the results are reported on data from NIST 2003 evaluations. Clear advantage of MMI over ML training is shown. Further improvements of acoustic LID are discussed: Heteroscedastic Linear Discriminant Analysis (HLDA) for feature de-correlation and dimensionality reduction and Ergodic Hidden Markov models (EHMM) for better modeling of dynamics in the acoustic space. The final error rate compares favorably to other results published on NIST 2003 data.

100 citations


Journal ArticleDOI
TL;DR: An inverse analysis based on the coupling of a stochastic simulation and an artificial neural network to obtain parameters of a computational model in order to achieve the best agreement with experimental data is proposed.

98 citations


Journal ArticleDOI
TL;DR: In this article, the linear thermal and hygric expansion coefficients of three lime plasters with pozzolanic admixtures are determined in the comparison with common lime plaster, and it can be concluded that all the analyzed lime-pozzolana plasters are suitable for an application in reconstruction of historical buildings.

Journal ArticleDOI
TL;DR: In this article, structural features of a South Moravian lignite and its humic acid (HA) fraction were studied by means of UV-vis, 1 H and solid-state 13 C NMR, and FT-IR spectroscopy in addition to basic elemental analyses.

Journal ArticleDOI
TL;DR: Nemcansky et al. as mentioned in this paper published a pioneering work on helical baffles in a shell-and-tube heat exchanger, paving the way to a major shift from a conventional understanding of baffles.
Abstract: Heat exchangers contribute significantly to many energy conversion processes. Applications range from power production, petroleum refining and chemicals, paper and pharmaceutical production, to aviation and transportation industries. A large percentage of world market for heat exchangers is served by the industry workhorse, the shell-and-tube heat exchanger. Recent developments in other exchanger geometries have penetrated in various industry applications; however, the shell-and-tube exchanger by far remains the industry choice where reliability and maintainability are vital. Over the years, significant research and development efforts are devoted to better understand the shell-side geometry. New geometries are introduced for performance enhancement and to improve reliability. The pioneering work published by J. Nemcansky et al. in the Trans. Institute of Chemical Engineers in May, 1990, on helical baffles paved the way to a major shift from a conventional understanding of baffles in a shell-and-tube heat...

Journal ArticleDOI
TL;DR: In this article, the tropospheric zenith delays from VLBI and GPS are in good agreement at the 3-7 mm level, while only small biases can be found for most of the stations, at Kokee Park (Hawaii, USA).
Abstract: In October 2002, 15 continuous days of Very Long Baseline Interferometry (VLBI) data were observed in the Continuous VLBI 2002 (CONT02) campaign. All eight radio telescopes involved in CONT02 were co-located with at least one other space-geodetic technique, and three of them also with a Water Vapor Radiometer (WVR). The goal of this paper is to compare the tropospheric zenith delays observed during CONT02 by VLBI, Global Positioning System (GPS), Doppler Orbitography Radiopositioning Integrated by Satellite (DORIS) and WVR and to compare them also with operational pressure level data from the European Centre for Medium-Range Weather Forecasts (ECMWF). We show that the tropospheric zenith delays from VLBI and GPS are in good agreement at the 3–7 mm level. However, while only small biases can be found for most of the stations, at Kokee Park (Hawaii, USA) and Westford (Massachusetts, USA) the zenith delays derived by GPS are larger by more than 5 mm than those from VLBI. At three of the four DORIS stations, there is also a fairly good agreement with GPS and VLBI (about 10 mm), but at Kokee Park the agreement is only at about 30 mm standard deviation, probably due to the much older installation and type of DORIS equipment. This comparison also allows testing of different DORIS analysis strategies with respect to their real impact on the precision of the derived tropospheric parameters. Ground truth information about the zenith delays can also be obtained from the ECMWF numerical weather model and at three sites using WVR measurements, allowing for comparisons with results from the space-geodetic techniques. While there is a good agreement (with some problems mentioned above about DORIS) among the space-geodetic techniques, the comparison with WVR and ECMWF is at a lower accuracy level. The complete CONT02 data set is sufficient to derive a good estimate of the actual precision and accuracy of each geodetic technique for applications in meteorology.

Journal ArticleDOI
TL;DR: The porosity, bulk density, compressive strength, bending strength, thermal conductivity, and specific heat capacity of a material based on alkali activated slag are determined at room temperature conditions for samples subjected to a thermal load up to 1200°C prior to the measurements and compared to reference material data.
Abstract: The porosity, bulk density, compressive strength, bending strength, thermal conductivity, and specific heat capacity of a material based on alkali activated slag are determined at room temperature conditions for samples subjected to a thermal load up to 1200°C prior to the measurements and compared to reference material data. The results are discussed using material characterization experiments, namely, X-ray diffraction analysis and scanning electron microscopy. Experimental results show a remarkable high-temperature resistance of the studied material, which after pre-heating to 1200°C exhibits mechanical properties comparable to the reference material not exposed to any thermal load.

Journal ArticleDOI
TL;DR: A method based on deterministic evaluation of measurement using the state observer based on the Lyapunov function is presented and has been proven in testing on a real ac induction machine.
Abstract: AC induction motors have become very popular for motion-control applications due to their simple and reliable construction. Control of drives based on ac induction motors is a quite complex task. Provided the vector-control algorithm is used, not only the rotor speed but also the position of the magnetic flux inside the motor during the control process should be known. In most applications, the flux sensors are omitted and the magnetic-flux phasor position has to be calculated. However, there are also applications in which even speed sensors should be omitted. In such a situation, the task of state reconstruction can be solved only from voltage and current measurements. In the current paper, a method based on deterministic evaluation of measurement using the state observer based on the Lyapunov function is presented. The method has been proven in testing on a real ac induction machine.

Journal ArticleDOI
TL;DR: In this article, the authors give formulas for solving the initial problem of linear discrete systems with constant square matrices with constant delay and a vector function f ( k ) with constant matrix delay.

Book ChapterDOI
29 Aug 2006
TL;DR: In this article, the verification of non-recursive C programs with dynamic linked data structures with possibly several next pointer selectors and with finite domain non-pointer data is considered, where the verification problem is considered to checking reachability of an error control line.
Abstract: We consider the verification of non-recursive C programs manipulating dynamic linked data structures with possibly several next pointer selectors and with finite domain non-pointer data. We aim at checking basic memory consistency properties (no null pointer assignments, etc.) and shape invariants whose violation can be expressed in an existential fragment of a first order logic over graphs. We formalise this fragment as a logic for specifying bad memory patterns whose formulae may be translated to testers written in C that can be attached to the program, thus reducing the verification problem considered to checking reachability of an error control line. We encode configurations of programs, which are essentially shape graphs, in an original way as extended tree automata and we represent program statements by tree transducers. Then, we use the abstract regular tree model checking framework for a fully automated verification. The method has been implemented and successfully applied on several case studies.

Journal ArticleDOI
TL;DR: A novel method utilising comprehensive two-dimensional liquid chromatography interfaced to electrospray ionisation time-of-flight mass spectrometry was developed for the determination of organic acids in atmospheric aerosols.

Journal ArticleDOI
TL;DR: In this paper, high-resolution vertical profiles of trace metals in freshwater sediments were obtained from the Upper Scheldt River (At Helkijn, Belgium) and the Leie River (at Warneton, located at the Belgian-French border).

Journal ArticleDOI
TL;DR: In this article, a femtosecond laser-based technique was used to analyze the spatial distribution of Fe within the leaves of two leaf samples and to identify the storage and trafficking of iron ions within different plant compartments.
Abstract: Using a femtosecond laser-based technique, spectro-chemical analysis of leaf samples is demonstrated. The study is exemplified for Fe. Standard reference concentrations - internal standards - of Fe in the two leaf samples were measured using the method of Relaxation Weighted Magnetic Resonance Imaging. From the spectra obtained using Laser Induced Breakdown Spectroscopy technique, spatial distribution of Fe within the leaf was identified. Thus, this technique could potentially be used as a complementary technique for identification of storage and trafficking of iron ions within different plant compartments. In particular, individual plant cells can be investigated without collateral damage with high spatial distribution.

Journal ArticleDOI
TL;DR: In this paper, the authors developed a method for construction of solutions of linear discrete systems with constant coefficients and with pure delay, which is expressed with the aid of a special function called the discrete matrix delayed exponential having between every two adjoining knots the form of a polynomial.
Abstract: The purpose of this contribution is to develop a method for construction of solutions of linear discrete systems with constant coefficients and with pure delay. Solutions are expressed with the aid of a special function called the discrete matrix delayed exponential having between every two adjoining knots the form of a polynomial. These polynomials have increasing degrees in the right direction. Such approach results in a possibility to express initial Cauchy problem in the closed form.

Journal Article
TL;DR: Major developments such as improvements in automatic segmentation, cross-domain model adaptation, inclusion of MLP based features, improvements in decoding, language modelling and vocal tract length normalisation, the use of a new decoder, and a new system architecture are described.
Abstract: We present the AMI 2006 system for the transcription of speech in meetings. The system was jointly developed by multiple sites on the basis of the 2005 system for participation in the NIST RT'05 evaluations. The paper describes major developments such as improvements in automatic segmentation, cross-domain model adaptation, inclusion of MLP based features, improvements in decoding, language modelling and vocal tract length normalisation, the use of a new decoder, and a new system architecture. This is followed by a comprehensive description of the final system and its performance in the NIST RT'06s evaluations. In comparison to the previous year word error rate results on the individual headset microphone task were reduced by 20% relative.

Journal ArticleDOI
TL;DR: Biofilm HPC values were higher at 35 degrees C than at 15 degrees C, with only slightly higher values on the metals, but 100-fold higher on PE and PEX, and adenosine triphosphate measurements confirmed the general trends observed by HPC.

Journal ArticleDOI
01 Oct 2006-Energy
TL;DR: In this paper, the authors presented a methodology for the development of an optimal solution to an integrated renewable energy retrofit in a large hospital building, where the design criteria include the possibility of utilizing renewable energy sources and providing a potential increase in heat production with additional heat being supplied to a nearby university campus.

Journal ArticleDOI
TL;DR: A generalisation of ARMC to the case of dealing with trees which arise naturally in a lot of modelling and verification contexts is proposed and it is shown on several examples that the methods given give us very encouraging verification results.

Journal ArticleDOI
TL;DR: In this paper, a free-volume viscosity model was used to predict the temperature-pressure-viscosity relationship of a squalane-lubricated circular contact.
Abstract: The free-volume viscosity model can accurately predict the temperature–pressure–viscosity relationship of lubricants. However, it is seldom used in elastohydrodynamic lubrication (EHL) simulation. This paper presents the application of the free-volume viscosity model in a Newtonian EHL simulation of a squalane-lubricated circular contact. Good agreement is observed between available experimental data and simulation results. The pressure–viscosity coefficients fit from viscometer data are also discussed. A recently developed definition of the coefficient is used to compare the coefficient value extracted from EHL film thickness interference measurements. Results indicate that the coefficient values from the curve fitting and EHL film thickness extraction agree well which has not been previously observed. Two factors help achieve this agreement: the new coefficient definition and smaller prediction error when using the Hamrock–Dowson formula in the cases studied. The effects of different pressure–viscosity relationships, including the exponential model, the Roelands model and the free-volume model, are investigated through an example with bright stock mineral oil. It is found that the real pressure–viscosity behavior predicted by the free-volume model yields a higher viscosity at the low-pressure area which results in a larger central film thickness. Therefore, due to use of the free-volume model, the present results are more consistent with experimental observations than previously reported numerical results.

01 Oct 2006
TL;DR: In this article, a numerical image processing method for visualization of corona images taken during total solar eclipses is presented, which is based on the use of adaptive filters inspired by human vision and the sensitivity of resulting images is thus very close to that of the human eye during an eclipse.
Abstract: A new numerical method of image processing suitable for visual- ization of corona images taken during total solar eclipses is presented This method allows us to study both small- and large-scale coronal structures that remain invisible on original images because of their very high dynamic range of the coronal brightness The method is based on the use of adaptive filters inspired by human vision and the sensitivity of resulting images is thus very close to that of the human eye during an eclipse A high precision alignment method for white-light corona images is also discussed The proposed method highly improves a widely used unsharp masking method employing a radially blurred mask The results of these numerical image processing techniques are illustrated by a series of images taken during eclipses of the last decade The method minimizes the risk of processing artifacts

Journal ArticleDOI
TL;DR: The thermal and thermooxidative properties of hyaluronic acid derivatives have been studied by means of differential scanning calorimetry and thermogravimetry as mentioned in this paper.
Abstract: The thermal and thermooxidative behavior of sodium salt of hyaluronic acid (HA) and its n-hexyl, n-decyl, n-tetradecyl and n-hexadecyl ether derivatives having an equal degree of substitution have been studied by means of differential scanning calorimetry (DSC) and thermogravimetry (TG). Derivatives were prepared by a substitution of H atom at the OH bound to the sixth C of N-acetyl-D-glucosamin of HA unit by n-hexyl, n-decyl, n-tetradecyl and n-hexadecyl chains. Both thermal and thermooxidative degradation of HA and derivatives resulted in multistep process. The main interest of this work was focused on processes occurring in the course of the first decomposition step. Experimental DSC data showed lower stability of derivatives and, remarkably lower heat evolution in comparison with original HA. On the other hand, TG measurement recorded lower mass loss for derivates which indicated appearance of new types of crosslinking reactions. Oxidative stability was evaluated by means of DSC that provided the induction period and the protection factor determination. Derivates showed remarkably lower stability in comparison with original HA; comparing each other, the highest oxidation stability showed n-decyl and n-tetradecyl derivates.

Journal ArticleDOI
01 Feb 2006
TL;DR: In this paper, the authors proposed a method to achieve very low N2O emissions from waste incineration, particularly for waste with higher nitrogen content (e.g., sewage sludge, leather, etc.).
Abstract: EU energy and environmental policy in waste management leads to increasing interest in developing methods for waste disposal with minimum emissions of greenhouse gases and minimum environmental impacts. From the point of view of nitrous oxide (N2O) emissions, waste incineration and waste co-combustion is very acceptable method of waste disposal. Two factors are important for attaining very low N2O emissions from waste incineration, particularly for waste with higher nitrogen content (e.g. sewage sludge, leather, etc.): temperature of incineration over 900°C and avoiding selective noncatalytic reduction (SNCR) de-NOx method based on urea. For reduction of N2O emissions retrofitting such plants to ammonia-based SNCR is recommendable. The modern selective catalytic reduction facilities for de-NOx at waste incineration plants are only negligible source of N2O.