Objectives. The development of contemporary models for the conversion of accents in foreign languages utilizes deep neural network architectures, as well as ensembles of neural networks for speech recognition and generation. However, restricted access to implementations of such models limits their application, study, and further development. Moreover, the use of these models is limited by their architectural features, which prevents flexible changes from being carried out in the timbre of the generated speech and requires the accumulation of context, leading to increased delays in generation, making these systems unsuitable for use in real-time multiuser communication scenarios. Therefore, the relevant task and aim of this work is the development of a method that generates native-sounding speech based on input accented speech material with minimal delays and the capability to preserve, clone, and modify the timbre of the speaker’s voice.
Methods. Methods for modifying, training, and combining deep neural networks into a single end-to-end architecture for direct speech-to-speech conversion are applied. For training, original and modified open-source datasets were used.
Results. The work resulted in the development of a real-time accent conversion method with voice cloning based on a non-autoregressive neural network. The model comprises modules for accent and gender detection, speaker identification, speech conversion, spectrogram generation, and decoding the resulting spectrogram into an audio signal. As well as demonstrating high accent conversion quality while maintaining the original timbre, the short generation times of the applied method make it acceptable for use in real-time scenarios.
Conclusions. Testing of the developed method confirmed the effectiveness of the proposed non-autoregressive neural network architecture. The developed model demonstrated the ability to work in real-time information systems in English.

Objectives. Despite the recent success of large language models, which are now capable of solving a wide range of tasks, a number of practical issues remain unsolved. For example, users of systems providing question answering (QA) services may experience a lack of commonsense knowledge and reasoning proficiency. The present work considers knowledge injection methods as a means of providing functional enhancements to large language models by providing necessary facts and patterns from external sources.
Methods. Knowledge injection methods leveraged in relevant QA systems are classified, analyzed, and compared. Self-supervised learning, fine-tuning, attention mechanism and interaction tokens for supporting information injection are considered along with auxiliary approaches for emphasizing the most relevant facts.
Results. The reviewed QA systems explicitly show the accuracy increase on the CommonsenseQA benchmark compared to pretrained language model baseline due to knowledge injection methods exploitation. At the same time, in general the higher results are related to knowledge injection methods based on language models and attention mechanism.
Conclusions. The presented systematic review of existing external knowledge injection methods for QA systems confirms the continuing validity of this research direction. Such methods are not only capable of increasing the accuracy of QA systems but also mitigating issues with interpretability and factual obsolescence in pretrained models. Further investigations will be carried out to improve and optimize different aspects of the current approaches and develop conceptually novel ideas.

Objectives. Pipelining is an effective method for increasing the clock frequency of digital circuits. At the same time, balancing the pipeline stages during circuit synthesis at the register transfer level does not yet guarantee a balanced topological implementation of such a pipeline in terms of signal propagation delays according to the selected technological basis. This is due to the specifics of the algorithms for placing and routing components of digital devices, which are not capable of optimizing solutions in a strict mathematical sense in an acceptable time. In practice, approaches for developing digital devices combine manual control of topological constraints that set general rules for placing components with automatic optimization for localized fragments of the circuit are used to obtain results close to optimal. Pipeline circuits are based on a simple connection diagram of individual stages to demonstrate the effect of using topological design constraints on their example. On the basis of pipeline structures, a number of algorithms can be implemented to effectively complement programmable processor devices and provide hardware acceleration of some tasks. The present work develops methodological recommendations for managing topological design constraints in the implementation of pipeline computing structures based on programmable logic devices (PLD) with field-programmable gate array (FPGA) architecture.
Methods. The work is based on accepted methods for designing and modeling digital systems.
Results. Based on the analysis, modifications to a 32-bit CORDIC transcendental function computation pipeline were developed. By adding design constraints on the placement of register groups corresponding to the pipeline stages a significant increase in the clock frequency can be achieved as compared to automatic placement to reduce the running time of the tracing algorithms. The resulting effect is systematically reproduced in several implemented versions of the pipeline. 
Conclusions. The presented recommendations can be used to control the clock frequency and number of stages of pipeline computing structures while simultaneously reducing the time of one iteration and routing of a module based on PLD with FPGA architecture.

Objectives. The active development of intelligent automatic control systems, which is associated with increasing requirements to the quality and accuracy of control systems of modern technical systems, requires the development of new approaches to their analysis and synthesis. A promising class of intelligent control devices is based on regulators that use fuzzy-logic inference technology. The purpose of this work is to develop a method for the complex synthesis of type-1 fuzzy regulator parameters on the basis of the Yakubovich circle criterion.
Methods. The proposed methodology is based on a consideration of fuzzy regulators in terms of the corresponding nonlinear transformation that support the use of methods derived from the theory of nonlinear automatic control systems. Analogs of the degrees of stability and oscillation are used as quality indicators. The synthesis of the parameters of the nonlinear transformation can be reduced to determining sufficient regions of absolute stability of the system with the shifted and extended Nyquist plot obtained using the Yakubovich circle stability criterion.
Results. In accordance with the theory of fuzzy sets and algorithms of fuzzy logical inference described by Takagi–Sugeno, the possibility of one-to-one correspondence of the nonlinear transformation and the parameters of an appropriately arranged knowledge base of the fuzzy controller is shown. A procedure proposed for synthesizing the parameters of the type-1 fuzzy regulator is aimed at ensuring complex requirements for the quality of the control system according to the degree of stability, the degree of oscillation, and steady-state mode accuracy. The effectiveness of the proposed technique, which guarantees the absolute stability not only of the equilibrium position but also of the processes, is confirmed by the results of model experiments.
Conclusions. The paper proposes a convenient engineering technique for determining the parameters of an intelligent controller constructed using fuzzy logic inference technology based on methods informed by automatic control theory. The convenience of using such indirect quality indicators as the degree of stability, the degree of oscillation, and accuracy in steady-state mode, is demonstrated. These indicators are explicable for developers of applied control systems.

Objectives. In the context ofmilitary operations, states face the threats of attacks by unmanned aerial vehicles (UAVs) on vulnerable assets, in particular, those pertaining to law enforcement agencies. Currently, there is no uniform or— more importantly—effective approach to detecting and suppressing certain types of UAVs, in particular, first person view (FPV) drones. The aim of the work is to develop modernized systems for the detection and suppression of enemy unmanned aerial vehicles and to justify their full-scale implementation in the service activities of law enforcement agencies.
Methods. The work used system-structural, comparative-legal, and measurement research methods along with analysis, observation, and field modeling. In addition, the research refers to generalized and systematized experience of using UAVs in combat conditions.
Results. The structure, basic tactical and technical characteristics are described according to the principle of operation of the identification complex for detecting and suppressing UAVs, having the ability to intercept analog radio signals carrying video information, which makes it possible to effectively detect and counteract enemy UAVs at a considerable distance. An algorithm of actions to be taken by law enforcement officers when using this complex is also developed and described. Proposals for creating a hardware and software system based on the complex with the possibility of spoofing a video stream are outlined.
Conclusions. The results of the study indicate the need to supply law enforcement agencies with an identification system for detecting and suppressing UAVs having the ability to intercept an analog radio signal carrying video information. The use of such a system across a wide area subject to UAV attacks will significantly improve the effectiveness of alerting all categories of employees and civilians, as well as contributing to the establishment of airspace control and improving the effectiveness of the fight against UAVs.

Objectives. In recent years, more and more attention has been paid in radar theory and practice to the development of multiple-input and multiple-output (MIMO) radar, which offers a number of advantages over traditional radar based on phased antenna arrays (PAAs). These include the possibility to flexibly view space and adapt to a changing signal-interference environment, etc. MIMO technology used in radar requires the emission of a probe signal in the form of a coherent system of orthogonal signals, each of which triggers its own emitter in the transmitting antenna array (AA). As a result, the specified target search area is simultaneously illuminated. Specific spatiotemporal processing (SSP) is used to collect signals from all directions in the irradiated zone at the receiver output. In this regard, the task of finding an SSP structure in MIMO radar that is optimal compared to the traditional approach becomes urgent. The study set out to synthesize the structure of SSP with single–channel reception in MIMO radar and compare the obtained structure and characteristics with those similar in traditional parallel-view radars based on multipath receiving radar.
Methods. The study is based on methods and principles of the theory of multibeam synthesized aperture antennas and methods for the synthesis of optimal Neiman–Pearson detectors based on the likelihood ratio.
Results. For a MIMO radar with AA for transmission and reception provided by a single weakly directional antenna, a split SSP was synthesized to form optimal pre-threshold statistics (PTS) of the detector against a background of white Gaussian noise. The obtained PTS is compared with a similar PTS in a traditional parallel space survey radar with a mirror structure.
Conclusions. It is shown that the detection quality indicators of the compared radars in the mirror construction are equivalent in the mode of parallel target search in the same spatial sectors.

Objectives. The work set out to investigate the possibility and effectivity of using a movable cylindrical defect with metal pins in the design of a photonic crystal to control the frequency response of a narrow-band filter in a rectangular waveguide having a cross-section of 23 × 10 mm in the X-band, as well as to determine the most effective methods for controlling frequency response.
Methods. A numerical simulation of the frequency response of the filter was carried out using the openEMS software package, which is based on Maxwell’s equations solved by the finite-difference time-domain method. The frequency response of the currently proposed and implemented filter construction in the X-band was further investigated in an experimental study.
Results. Numerical simulation shows that a resonant transmission peak in the stopband of the frequency response can be caused to appear by introducing a movable cylindrical defect having two metal pins into the center of a photonic crystal structure. In addition, the position of this peak on the frequency response can be effectively controlled by rotating the cylindrical defect around its axis. If the position of the defect remains unchanged, an increase in the frequency of the transmission peak occurs as a result of decreasing the period of the photonic crystal. However, the frequency of this resonant transmission peak is most strongly influenced by changes in the size of holes in the photonic structure. These changes can be used to control both the position and shape of the transmission peak, as well as the overall frequency response. At the same time, the difference in transmission remains practically unchanged when the cylinder rotates around its axis. The simulation results were confirmed by the data of an experimental study of the frequency response of photonic crystals made from PETG plastic using 3D printing technology.
Conclusions. The proposed, designed, and manufactured experimental samples of narrow-band filters in the X-band based on a photonic crystal demonstrated reliably variable transmission values and the possibility of controlling the resonant peak frequency and thus the entire frequency response, including operational control. This makes them very promising for practical use in radio-electronic equipment.

Objectives. Sections of microstrip lines having finite length are widely used to develop integrated circuits and microwave devices for various purposes, such as power dividers, directional couplers, attenuators, and filters. In particular, low-pass filters in the microwave range are comprised of a cascade connection of regular sections of microstrip lines having various geometric parameters. However, modern approaches to calculating microwave filters using commercial software require large computational and time-consuming resources, especially when carrying out electrodynamic analysis of microstrip lines. The work set out to develop an algorithm and a method for calculating filters using a projection approach to the electrodynamic analysis of microstrip lines that reduces the time required to calculate characteristics of microwave filters while maintaining high accuracy of the obtained results.
Methods. The proposed projection approach to the electrodynamic analysis of a microstrip line can be used to rapidly and accurately calculate the main electrodynamic parameters of retardation coefficient and wave impedance across a wide range of changes in the geometrical parameters of the line, as well as its dielectric constant and frequency.
Results. Formulas obtained on the basis of analytical expressions for calculating the electrodynamic parameters of a microstrip line are used to describe the nature of changes in the elements of the scattering matrix of multistage low-pass filters in a given frequency band. A developed computer program was used to calculate the values of the elements of the low-pass filter scattering matrix across a wide range of substrate dielectric constant and frequency parameters. The obtained results were compared with the characteristics of filters calculated using commercial software.
Conclusions. The proposed approach to calculating the electrodynamic parameters of microstrip lines and consequent elements of the scattering matrix of multistage low-pass filters can significantly reduce the calculation time while achieving a sufficiently high accuracy of the obtained results to significantly reduce labor costs when calculating microwave filters in engineering practice.

Objectives. Microsystem engineering is currently receiving a great deal of research attention due to the very wide scope of application of its various elements. The present study of the development and creation of modern gyroscopes based on microelectromechanical systems (MEMS gyroscopes) analyzes the risks associated with the technological aspects of their production and identifies promising areas for further development both of MEMS gyroscopes themselves and the technologies used to manufacture them.
Methods. A detailed analysis of existing scientific publications, analytical reviews, and other available sources on MEMS gyroscopes and current trends in the field of microoptoelectromechanical technologies and ferroelectric films was carried out.
Results. A brief description of the design solutions of modern MEMS gyroscopes and their integration into mechatronic systems is presented. The production technologies of MEMS gyroscopes and specifics of the technological equipment used are considered. A separate section discusses the configuration and calibration aspects of these devices. Promising directions for the development of MEMS gyroscopes with an emphasis on the use of microoptoelectromechanical converters and ferroelectric films are highlighted.
Conclusions. Based on the analysis, the prospects for the development of MEMS gyroscopes are shown, despite the existing technological challenges. It is noted that new physical principles and unique technologies can contribute tothe emergence ofnew types ofMEMS gyroscopes using micro-optoelectromechanical converters and ferroelectric films. This, in turn, opens up new horizons for future developments in this area. The necessity of developing new production technologies and specialized equipment to improve the quality of MEMS gyroscopes is demonstrated.

Objectives. The work set out to study the spectra of the magnetorefractive effect (MRE) in the cobalt–silicon (Co–Si) nanocomposite, taking into account the contribution of the size effect(SE), and to compare the results obtained by varying the parameters of the SE. The presented approaches to investigating the magnetooptical properties of nanocomposites, which are relevant for the practical application of nondestructive testing methods, have the potential to significantly increase the efficiency of their use in various fields, including spintronics and optics.
Methods. Computer modeling approaches based on the Bruggeman approximation are used to model the examined structure as a medium with effective properties.
Results. MRE spectra obtained within the framework of the modeling fell within the range of 0.5–3.5 eV. The modeling was carried out for MRE both with and without taking into account the semiclassical size effect. The resultant modeling of the spectral dependencies of the MRE is based on the example of a Co–Si nanocomposite at different cobalt particle sizes and form factors. The influence of size effects on the form of the MRE spectra is confirmed. The reliability of the methods is confirmed by a comparison of the obtained results with empirical data. The value of the obtained results consists in the good agreement of all the calculated parameters of the discussed nanocomposite and the form of the spectral dependencies of the MRE with the results of various experiments.
Conclusions. The confirmation that both the size and form factor of granules have a significant impact on the appearance of the MRE spectra raises the prospect of developing promising nanocomposite properties at particular particle sizes. The presented results highlight the possibility of optimizing the material characteristics to improve sensitivity in magnetic sensors and noncontact devices for studying nanostructures.