738 research outputs found
Language-independent, neural network-based, text-to-phones conversion
No full textA speech synthesizer based on an artificial neural network (ANN) is being developed for application to deeply embedded systems for language-independent speech commands on hands-free interfaces. A feed-forward, backpropagation, artificial neural network has been trained for this purpose using a custom-developed, regular expression-based, text-to-phone transcription engine to generate training patterns. Initial experimental results show the expected properties of language independence and in-system learning capability of this approach. The ANN demonstrates the capacity to generalize and map the words missing at training time, as well as to reduce contradictions related to different pronunciations for the same word
A Language-Independent Neural Network-Based Speech Synthesizer
No full textThe applicability of soft computing to implementing text-to-speech conversion is subject to debate. Using neural networks for phoneme-level, text-to-speech conversion has several advantages over hard computing. Soft computing’s capacity to generalize makes it possible to map words missing from the database, as well as to reduce contradictions related to different pronunciations for the same word.
Neural networks have been shown to optimally solve a large class of applied pattern-matching problems, but very little research has been done to match the requirements of pattern generation in machine-to-human interaction.
An artificial speech synthesizer based on neural networks is being developed for application to deeply embedded systems for language-independent speech commands on hands-free interfaces. A feed-forward, backpropagation artificial neural network has been trained for this purpose using a custom-developed, regular expression-based, text-to-phones transcription engine to generate training patterns. Initial experimental results show the expected properties of language independence and in-system learning
Implementation of a new bi-directional solar modelling method for complex facades within the ESP-r building simulation program
No full textThis paper provides an overview of a new method for modelling the total solar energy transmittance. It is implemented in the ESP-r building simulation program to model complex façades such as double glazed façades with external, internal or integrated shading devices. This new model has been validated and tested for several cases. The new model required changes to the solar control simulation algorithm and the user interface, so a new “Advanced optics menu” was also introduced into ESP-r. The paper presents the interface development and application of the new technique to different simulation configurations (especially different complex façades with shading devices) in a standard office building
Synthesis on switching lattices of Dimension-reducible Boolean functions
Get PDFIn this paper we study the switching lattice synthesis of a special class of regular Boolean functions called D-reducible functions. D-reducible functions are functions whose points are completely contained in an affine space A strictly smaller than the whole Boolean cube 0, 1n. The D-reducibility of a function can be exploited in the lattice synthesis process: the idea is to independently find lattice implementations for the characteristic function of the subspace A and for the projection of onto A, and to compose them in order to construct the lattice for. The overall lattice area can be further reduced exploiting the peculiar structure of the affine subspaces of 0, 1n. To this aim, we propose a method for implementing compact lattice representations of affine subspaces whose characteristic function is represented by the product of single literals and EXOR factors of two literals. The experimental results validate the proposed approach
Advances in the application of the Principal Static Wind Loads: A large-span roof case
No full textIn this paper, a procedure to define a small set Multi-target Equivalent Static Wind Loads (M-ESWLs) that allows for the verification of a large set of structural elements of a stadium roof is presented and applied to a real case. The methodology, based on the combination of the so-called Principal Static Wind Loads (PSWLs), is aimed at reducing the subjectivity of the designer in the identification of the most challenging structural conditions induced by wind actions, and therefore it allows to simplify and speed up the wind-resistant design process of roof structures.Starting from wind tunnel tests on a scaled rigid model of a stadium, the dynamic response of its roof is computed by loading the finite element model of the structure with the time histories of the spatially averaged net pressures acting on its surface. The time histories of the internal forces acting in a set of structural members are reconstructed, and the target envelope is built by identifying, for each of them, a maximum and a minimum value through a statistical analysis of the time series. For each of these extreme conditions an ESWL is identified, PSWLs are computed and combined by means of a genetic algorithm, and the set of M-ESWLs, able to match the envelope, is defined. To limit the number of M-ESWLs, a nonlinear constrained optimization is implemented in the PSWLs combination, also introducing the possibility to over-dimension some structural members.It is concluded that the application of this methodology allows the designer to select a limited number of M-ESWLs that can be used in an automated way in order to verify the response of a defined set of structural elements
Development of a new vertical angle selective façade for solar control
No full textMany recent researches attests that the 40% of the European energy consumption it’s due to the building. In particular the 80% of building energy consumption currently occurs during service-life. The energy demand for cooling is raising day by day (especially in Mediterranean climates.
But it’s also attested that all the buildings, residential and commercial, have an enormous potential for improvement the energy efficiency (think about the standard passive-house in Germany and Austria) and of energy systems integration, like pv-systems or solar cooling systems, to produce energy from sun. This is particularly true for non-residential sector and high-rise buildings. The penetration in residential buildings (solar domestic hot water and pv-systems), in particular in single family house, it’s much better in Germany and Central Europe, where the incentives from the States cover a good percentage of the installation costs. Other matter is the integration into high or tall buildings where the only place to integrate the systems is the façade.
In facade construction, in particular for high-rise buildings, the portion of the glazed area of the façade rises continuously. The architects like very much transparency, the sun and the contact with the city and the context. This causes the positive effect of higher daylighting for the internal space and reduction of energy consumption for artificial lighting. But, on the other hand, it produces the risk of overheating, in particular during summer period and/or high cooling loads For that reason sun protection (with solar control systems) is necessary.
It’s also very common in Central or North Europe to control the solar gains with external shading systems (for example venetian blinds). This solution has some advantages for residential house (family houses), because of the low cost for installation, but on the other hand venetian blinds doesn’t produce energy (no electricity production) and are not good for high-rise buildings and other windy locations.
The new angle selective façade combines in one element, three important tasks: solar protection, glare protection and integrated pv-system for electricity generation. These three elements as are completely integrated in the function of the façade doesn’t reduce the architectural goal of the glazed façade and the view from the interior to the exterior is guaranteed.
The new angle-selective-façade system is a static seasonal sun protection, which can be produced using the usual production technologies of windows and glazing units. The system is currently optimised in the Fraunhofer ISE in collaboration with the corresponding author who is a PHD student at Politechnico di Milano. In the publication we will describe the goals and the concept of the new façade element
Composition of switching lattices for regular and for decomposed functions
Get PDFMulti-terminal switching lattices are typically exploited for modeling switching nano-crossbar arrays that lead to the design and construction of emerging nanocomputers. Typically, the circuit is represented on a single lattice composed by four-terminal switches. In this paper, we propose a two-layer model in order to further minimize the area of regular functions, such as autosymmetric and D-reducible functions, and of decomposed functions. In particular, we propose a switching lattice optimization method for a special class of “regular” Boolean functions, called autosymmetric functions. Autosymmetry is a property that is frequent enough within Boolean functions to be interesting in the synthesis process. Each autosymmetric function can be synthesized through a new function (called restriction), depending on less variables and with a smaller on-set, which can be computed in polynomial time. In this paper we describe how to exploit the autosymmetry property of a Boolean function in order to obtain a smaller lattice representation in a reduced minimization time. The original Boolean function can be constructed through a composition of the restriction with some EXORs of subsets of the input variables. Similarly, the lattice implementation of the function can be constructed using some external lattices for the EXORs, whose outputs will be inputs to the lattice implementing the restriction. Finally, the output of the restriction lattice corresponds to the output of the original function. Experimental results show that the total area of the obtained lattices is often significantly reduced. Moreover, in many cases, the computational time necessary to minimize the restriction is smaller than the time necessary to perform the lattice synthesis of the entire function. Finally, we propose the application of this particular lattice composition technique, based on connected multiple lattices, to the synthesis on switching lattices of D-reducible Boolean functions, and to the more general framework of lattice synthesis based on logic function decomposition
- …
