The paper investigates possibility of creating a personal voice model using transcribed speech samples of a specified speaker. The paper presents a practical way of building such speech model and some experimental results of applying the model to voice conversion. The model uses an artificial neural network organized as autoencoder that establishes correspondence between space of speech parameters and space of possible phonetic states, unified for any voice.
State-of-the-art models of artificial intelligence are developed in the black-box paradigm, in which sensitive information is limited to input-output interfaces, while internal representations are not interpretable. The resulting algorithms lack explainability and transparency, requested for responsible application. This paper addresses the problem by a method for finding Osgood’s dimensions of affective meaning in multidimensional space of a pre-trained word2vec model of natural language. Three affective dimensions are found based on eight semantic prototypes, composed of individual words. Evaluation axis is found in 300-dimensional word2vec space as a difference between positive and negative prototypes. Potency and activity axes are defined from six process-semantic prototypes (perception, analysis, planning, action, progress, and evaluation), representing phases of a generalized circular process in that plane. All dimensions are found in simple analytical form, not requiring additional training. Dimensions are nearly orthogonal, as expected for independent semantic factors. Osgood’s semantics of any word2vec object is then retrieved by a simple projection of the corresponding vector to the identified dimensions. The developed approach opens the possibility for interpreting the inside of black box-type algorithms in natural affective-semantic categories, and provides insights into foundational principles of distributive vector models of natural language. In the reverse direction, the established mapping opens machine-learning models as rich sources of data for cognitive-behavioral research and technology.
One of the solutions to the problem of spatio-temporal data anisotropy is the use of a multilevel system of digital twins based on the corresponding industry models and the updated archive data base. The application of this approach has successfully proved itself in information systems for monitoring the parameters of the geomagnetic field and its variations, providing spatio-temporal interpolation of geomagnetic data with an accuracy of 0.81 nT in magnetically quiet periods. At the same time, the problem of information interaction between the levels of the system of digital twins remained unresolved, which is greatly aggravated by the constantly growing volume of data and their heterogeneous nature. The paper proposes a solution to the indicated problem by means of a formalized mechanism for packaging space-time information, in which the identification of data sources is performed on the basis of a hierarchical binary tokenization system. In addition, the proposed software implementation of such an approach is considered, a distinctive feature of which is the combination of traditional clientserver and innovative serverless architectures to implement a highly loaded reactive web application for working with analyzed data. The main stages of the implementation of information interaction are highlighted and programmatically formalized - from obtaining initial information from its sources to verifying data, analyzing them, processing and forming the output information flow of the system. The results of the computational experiments carried out on the example of the problem of monitoring the parameters of the Earth's magnetic field and its variations confirmed the effectiveness of the proposed solutions, expressed both in increasing the reactivity of web-based applications and in increasing the computational speed of formation and filling of information storages that aggregate information from distributed heterogeneous sources.
The purpose of the study is to select the optimal conditions for collecting non-coordinate information about a spacecraft by a space optical-electronic means at the time objects pass the vicinity of the points of the minimum distance between their orbits. The quantitative indicator is proposed that characterize the measure of the possibility of obtaining non-coordinate information about space objects with the required level of quality. The arguments of the function characterizing the indicator are the distance between spacecraft; their relative speed; phase angle of illumination of a spacecraft by the Sun in relation to the optical-electronic means; the length of the time interval during which both objects are in the vicinity of the point of a minimum distance between their orbits. The value of the indicator is computed by solving three particular research problems. The first task is to search for neighborhoods that include the minimum distances between the orbits of the controlled spacecraft and optical-electronic means. To solve it, a fast algorithm for calculating the minimum distance between orbits used. Additionally, the drift of the found neighborhoods is taken into account on the time interval up to 60 hours. The second task is to estimate the characteristics of motion and the conditions of optical visibility of a controlled spacecraft in the vicinity of the minimum points of the distance between the orbits of spacecraft. The solution to this problem is carried out by using the SGP4 library of space objects motion forecast. The third task is justification and calculation of an index characterizing the measure of the possibility of obtaining an optical image of a spacecraft for given conditions of optical visibility. To solve the problem, the developed system of fuzzy inference rules and the Mamdani algorithm is used. The presented method is implemented as a program. In the course of a computational experiment, an assessment was made of the possibility of obtaining non-coordinate information on low-orbit and geostationary space objects. The proposed indicator provides an increase in the efficiency of the procedure for collecting non-coordinate information about space objects by choosing the most informative alternatives for monitoring space objects from the available set of possible observations at a given planning interval for collecting information about space objects.
The development of methodical and mathematical apparatus for formation of a set of diagnostic parameters of complex technical systems, the content of which consists of processing the trajectories of the output processes of the system using the theory of functional spaces, is considered in this paper. The trajectories of the output variables are considered as Lebesgue measurable functions. It ensures a unified approach to obtaining diagnostic parameters regardless a physical nature of these variables and a set of their jump-like changes (finite discontinuities of trajectories). It adequately takes into account a complexity of the construction, a variety of physical principles and algorithms of systems operation. A structure of factor-spaces of measurable square Lebesgue integrable functions, ( spaces) is defined on sets of trajectories. The properties of these spaces allow to decompose the trajectories by the countable set of mutually orthogonal directions and represent them in the form of a convergent series. The choice of a set of diagnostic parameters as an ordered sequence of coefficients of decomposition of trajectories into partial sums of Fourier series is substantiated. The procedure of formation of a set of diagnostic parameters of the system, improved in comparison with the initial variants, when the trajectory is decomposed into a partial sum of Fourier series by an orthonormal Legendre basis, is presented. A method for the numerical determination of the power of such a set is proposed.
New aspects of obtaining diagnostic information from the vibration processes of the system are revealed. A structure of spaces of continuous square Riemann integrable functions ( spaces) is defined on the sets of vibrotrajectories. Since they are subspaces in the afore mentioned factor-spaces, the general methodological bases for the transformation of vibrotrajectories remain unchanged. However, the algorithmic component of the choice of diagnostic parameters becomes more specific and observable. It is demonstrated by implementing a numerical procedure for decomposing vibrotrajectories by an orthogonal trigonometric basis, which is contained in spaces. The processing of the results of experimental studies of the vibration process and the setting on this basis of a subset of diagnostic parameters in one of the control points of the system is provided.
The materials of the article are a contribution to the theory of obtaining information about the technical condition of complex systems. The applied value of the proposed development is a possibility of their use for the synthesis of algorithmic support of automated diagnostic tools.
. As is known, today the problem of geomagnetic field and its variations parameters monitoring is solved mainly by a network of magnetic observatories and variational stations, but a significant obstacle in the processing and analysis of the data thus obtained, along with their spatial anisotropy, are omissions or reliable inconsistency with the established format. Heterogeneity and anomalousness of the data excludes (significantly complicates) the possibility of their automatic integration and the application of frequency analysis tools to them. Known solutions for the integration of heterogeneous geomagnetic data are mainly based on the consolidation model and only partially solve the problem. The resulting data sets, as a rule, do not meet the requirements for real-time information systems, may include outliers, and omissions in the time series of geomagnetic data are eliminated by excluding missing or anomalous values from the final sample, which can obviously lead to both to the loss of relevant information, violation of the discretization step, and to heterogeneity of the time series. The paper proposes an approach to creating an integrated space of geomagnetic data based on a combination of consolidation and federalization models, including preliminary processing of the original time series with an optionally available procedure for their recovery and verification, focused on the use of cloud computing technologies and hierarchical format and processing speed of large amounts of data and, as a result, providing users with better and more homogeneous data.
We consider the problem of community detection for the graph which is a fragment of the academic Web. The nodes of the graph are the sites of the scientific organizations, and its arcs are hyperlinks. We propose a new approach based on the methods of coalition game theory to derive the Nash-stable coalition partition. This is determined by a function of preferences for any pair of vertices in the graph. The problem of finding a stable partition is connected with finding a maximum of potential function. The algorithm for searching stable partitioning and evaluating its complexity is presented. The proposed method was compared with two well-known methods of finding communities. The efficiency of the new method is demonstrated on the fragment of the Web which consists of the official sites of the Siberian and Far East branches of RAS.
Smart space is a service orientated infrastructure for knowledge sharing between devices. This paper describes a smart space creation process based on integration of Smart-M3 platform with DD-WRT-based device. Smart-M3 is an open source platform which implements the concept of smart space. Wi-Fi router was chosen as a DD-WRT-based device, which allows using it for smart space organization and provides wireless connection between devices at the same time. This method simplifies deployment of scenarios with several participants. The paper describes a process of compilation and installation of Smart-M3 platform to DD-WRT operating system. Testing of the main Smart-M3 operations showed that a smart space, organized in this way, can be used for considered scenarios. The authors have developed “Smart-M3 Control Panel” web-service which allows users to control the Smart-M3 platform by a graphical web interface. User of “Smart-M3 Control Panel” can view a current status of the platform; launch, stop or reload the platform; view information storage; download log files and change startup options. SocketIO protocol was used for user interaction with a web service.
An improvement of existing navigation algorithms for a generic polygonal linkage is presented. Our algorithm constructs a path between two arbitrary configurations of a polygonal linkage. This path contains att most eight steps
The paper presents the results of studies of a double-valued and a multi-valued elements stock (ES) for digital structures, including robots and flying vehicles, that operates by using linear transformations of current logic signals. A goal of the studies is to develop methods of synthesis and analog circuits engineering solutions of the alternative digital ES with the improved (as compared to the traditional elements stock) technical, technological and operational characteristics.
The article discusses a number of problems of space cybernetics connected with the optimum process control of informational interaction of a spacecraft with the surface of the Earth. A spacecraft is regarded as an informational active mobile plant, i.e. as a complicated mobile system supplied with necessary devices for the realization of informational interaction with the enclosing physical medium and the corresponding necessary onboard resource. It is shown that these problems are reduced to the problems of optimum programmed control by some special differential dynamic system in a Hilbert space of conditions. For a solution of the specified problems, the paper uses L.S. Pontrjagin's expanded maximum principle and the general concept of Lagrange.
A survey of smart space prototypes intended for scientific and educational meetings and facilitated by means of automatic speech recording is represented in the paper. Analysis of the used audio visual signal processing means and realized user services allowed us to propose a classification of smart space prototypes. The peculiarities of the developed smart meeting room and its distinction features from the considered prototypes are described.
The paper provides an analytical review of the invited talks by leading foreign and domestic experts in the security of computer networks, presented at the 2nd International Workshop “Scientific Analysis and Policy Support for Cyber Security” (SA&PS4CS 2012), held in St. Petersburg October 20, 2012. The following foreign scientists had presentations: V. Skormin (USA), F. Martinelli (Italy), V. Oleshchuk (USA), R. Rieke (Germany), A. Hutchison (RSA), L. Khan (USA), P. Laskov (Germany), and S. Mjølsnes (Norvey). The following Russian specialists were invited: V. Mayorov, A. Svistunov, R. Yusupov, A. Grusho, P. Zegzhda, A. Smirnov, A. Zemtsov, and I. Kotenko. The main topics of the workshop’s presentations were detection, recognition and identification of various types of malicious activity, responding to attacks and intrusions in cyberspace, including information operations at the national level, the identification of new promising technologies, techniques, methods and means of cooperation ensuring in the field of security policies support in cyberspace.
Methods and techniques of software design as one of the important stages of software development are described in the paper. The method of software design with using of UML with Petri nets for analyzing of dynamic properties of set UML diagrams is described. Authors offer improved method of using of integration of UML diagrams and Petri nets. The offered method was used for designing of software of automated process control system (APCS) of pumping station: designing of use case, class, object diagrams and sequences diagram that was transformed to Petri net with help of formal rules. Some incorrect states that occurred after pumps enabling/disabling by operator were identified by analysis of Petri net. Reachability tree of the system was gotten by analysis of the Petri net (the value of the tree is about 10 6 of nodes). Testing of offered system was showed on example of APCS of pump station.
In the article the mathematical model of informational interaction of the spacecraft with a surface of the Earth is considered. In a basis of construction of model lies proposed of the author the concept of active movable object as the complicated movable system intended for informational, energy or material interaction with an ambient physical environment or with other similar systems. It is shown that the corresponding model can be presented in the form of Fredholm integral operator mapping set of elements of a Hilbert space of controls (a class of admissible control actions) in a Hilbert space of information states. Investigated the properties of this operator and the corresponding reachability sets in space of information states. Considered a simplified variant of the proposed mathematical model - for interaction with the discrete environment (the isolated sources of information).
The article sets out the methods proposed for solving the generalized problem eigenvalues and vectors for singular matrix bundles, occurring in the important applied problems of different branches of knowledge.
This paper discuss the problem of personal adjustment of smart room devices and forming of a user profile based on processing of multichannel audio and video streams, which register of the current situation and meeting participants behavior in the meeting room. Estimation of preference of device usage, user interface, participant role and their activity during meeting allows us to automate the processes of smart room preparation as well as manage multimedia presentation and record devices during events. 212 records were made during several meetings in the smart room with the help of the developed system of audio and video speaker localization. The accumulated experimental data allowed us to estimate the places in the room, from which the participants asked questions most of the time. The accuracy of camera pointing on speaker in the presentation zone as well as in the rows of sits estimated by participant’s face size and its position in frame during whole recording equals 90% approximately.
The software ability to analyze current operating conditions, including a current state of a user, a physical environment, computing resources, as well as dynamically adaptation of the scenario of interaction to the user, is a one of the main requirements in the development of smart space prototypes. Control of a set of software and hardware modules involved in the smart space is difficult with increasing of tasks and number of users, so mathematical and software tools, which implement control of distributed modules, are required. The paper describes the structure of the model for distributed modules control in a smart meeting room as well as used multimodal interfaces for natural human-computer interaction.
Equilibrium shapes of a planar polygonal linkage are cyclic configurations. We describe all possible sets of cyclic configurations for pentagonal linkages.
The employing of sound source localization methods allows to evaluate location and head orientation of a speaker in a room. At present such systems are popular at development of intelligent support systems for smart meeting rooms. In this paper, a set of the metrics for performance evaluation of sound source localization systems as well as their integration with video monitoring systems are analyzed. Accuracy estimation of speaker positions, located on 32 chairs was carried out in the developed smart meeting room
Intelligent system for conference management automation is an interactive system for conferencing in a room equipped with two screens and facilities to project slides and other information on them. The first screen is used to display the speaker’s presentation, and the second screen is used for conference program visualization. The system is based on the Smart-M3 platform developed by the Nokia Research Center. Personal mobile devices (communicators, mobile phones, and personal computers) are used for interaction between users and the system. The appropriate software is installed on these devices. This article introduces the architecture and ontologies of the system's main components.
Current activities of the scientific-educational center “Intelligent space technologies » founded in 2010 for supporting investigations, coordination of integrated innovation projects, efficient production of scientific-educational personnel and keeping of youth in science and education areas are outlined.
Association rule search is one of the ever increasing areas of the intelligent data analysis and data mining. Unfortunately, traditional approaches of this area often are not capable to cope with new challenging problems emerging while dealing with new classes of modern applications. The latter require new viewpoint on methodology and technology of association analysis where "classical" ones fail. For efficient solution of emerging tasks of the association analysis, the paper proposes "non–classical" model of probabilistic space and its fragment called "sub–defined probabilistic space". While using algebraic view, the probabilistic models used are defined in terms of normalized Boolean algebra and lattices. Such a probabilistic model made it possible to cope with several challenging association analysis tasks. Between them, the proposed algorithm is capable of search for rare but "strong" association rules, mining negative rules of any forms and mining cause consequence rules. All these tasks are solved within the same framework called associative (algebraic) Bayesian network. The basic algorithm is demonstrated by simple case study, although the algorithm and corresponding software developed for this purpose were validated on an application of real life scale.
1 - 23 of 23 items