Search Results (344)

Emotion recognition in speech is essential for enhancing human–computer interaction (HCI) systems. Despite progress in Bangla speech emotion recognition, challenges remain, including low accuracy, speaker dependency, and poor generalization across emotional expressions. Previous approaches often rely on traditional machine learning or basic deep learning models, struggling with robustness and accuracy in noisy or varied data. In this study, we propose a novel multi-stream deep learning feature fusion approach for Bangla speech emotion recognition, addressing the limitations of existing methods. Our approach begins with various data augmentation techniques applied to the training dataset, enhancing the model’s robustness and generalization. We then extract a comprehensive set of handcrafted features, including Zero-Crossing Rate (ZCR), chromagram, spectral centroid, spectral roll-off, spectral contrast, spectral flatness, Mel-Frequency Cepstral Coefficients (MFCCs), Root Mean Square (RMS) energy, and Mel-spectrogram. Although these features are used as 1D numerical vectors, some of them are computed from time–frequency representations (e.g., chromagram, Mel-spectrogram) that can themselves be depicted as images, which is conceptually close to imaging-based analysis. These features capture key characteristics of the speech signal, providing valuable insights into the emotional content. Sequentially, we utilize a multi-stream deep learning architecture to automatically learn complex, hierarchical representations of the speech signal. This architecture consists of three distinct streams: the first stream uses 1D convolutional neural networks (1D CNNs), the second integrates 1D CNN with Long Short-Term Memory (LSTM), and the third combines 1D CNNs with bidirectional LSTM (Bi-LSTM). These models capture intricate emotional nuances that handcrafted features alone may not fully represent. For each of these models, we generate predicted scores and then employ ensemble learning with a soft voting technique to produce the final prediction. This fusion of handcrafted features, deep learning-derived features, and ensemble voting enhances the accuracy and robustness of emotion identification across multiple datasets. Our method demonstrates the effectiveness of combining various learning models to improve emotion recognition in Bangla speech, providing a more comprehensive solution compared with existing methods. We utilize three primary datasets—SUBESCO, BanglaSER, and a merged version of both—as well as two external datasets, RAVDESS and EMODB, to assess the performance of our models. Our method achieves impressive results with accuracies of 92.90%, 85.20%, 90.63%, 67.71%, and 69.25% for the SUBESCO, BanglaSER, merged SUBESCO and BanglaSER, RAVDESS, and EMODB datasets, respectively. These results demonstrate the effectiveness of combining handcrafted features with deep learning-based features through ensemble learning for robust emotion recognition in Bangla speech. Full article

For high-voltage signal detection applications, an auto-zero and chopper operational amplifier (OPA) is proposed in this paper. With the auto-zero and chopper technique, the OPA adopts an eight-channel Ping-Pong mechanism to reduce the high-frequency ripple and glitch generated by chopper modulation. The main transconductor effectively suppresses low-frequency noise and offset by combining input coarse and output fine auto-zero. A common-mode voltage tracking circuit is presented to ensure constant gate-source and gate-substrate voltages of the chopper, which reduces the charge injection caused by threshold voltage drift of their transistors and improves output signal resolution. The OPA is implemented using CMOS 180 nm BCD process. The post-simulation results show that the unit gain bandwidth (UGB) is 2.5 MHz and common-mode rejection ratio (CMRR) is 137 dB when the power supply voltage is 5–55 V. The noise power spectral density (PSD) is 6.6 nV/√Hz, and the offset is about 47 µV. The overall circuit consumes current of 960 µA. Full article

This paper introduces a specific matrix spectral problem involving four potentials and derives an associated soliton hierarchy using the zero-curvature formulation. The bi-Hamiltonian formulation is derived via the trace identity, thereby establishing the hierarchy’s Liouville integrability. This is exemplified through two systems: generalized combined NLS-type equations and modified KdV-type equations. Owing to Liouville integrability, each member of the hierarchy admits a bi-Hamiltonian structure and, consequently, possesses infinitely many symmetries and conservation laws. Full article

A multi-objective genetic algorithm is tailored to optimize the design of a wavelength interleaver/deinterleaver device. An interleaver combines data streams from two physical channels into one. The deinterleaver does the opposite job. The WDM requirements for this device include channel spacing of 50 GHz, channel bandwidth of 20 GHz, free spectral range of 100 GHz, maximum channel dispersion of 30 ps/nm, and maximum crosstalk of −23 dB. The challenges for the optimization process include the lack of a closed-form expression for the device performance and the trade-off between the conflicting performance parameters. So, for this multi-objective problem, the proposed approach maneuvers to find a compromise between the performance parameters within a few minutes, saving the designer the laborious design process previously proposed in the literature, which relies on visually inspecting the Z-plane for the dynamics of the transmission poles and zeros. Designs of better performance are achieved, with fewer ring resonators, a channel dispersion as low as 1.6 ps/nm, and crosstalk as low as −30 dB. Full article

The wave spectrum, as a key statistical feature describing wave energy distribution, is crucial for understanding wave propagation mechanisms and supporting ocean engineering applications. This study, based on ERA5 reanalysis spectrum data, proposes a model combining CNN and xLSTM for rapid gridded wave spectrum prediction over the Bohai and Yellow Seas domain. It uses 2D gridded spectrum data rather than a spectrum at specific points as input and analyzes the impact of various input factors at different time lags on wave development. The results show that incorporating water depth and mean sea level pressure significantly reduces errors. The model performs well across seasons with the seasonal spatial average root mean square error (SARMSE) of spectral energy remaining below 0.040 m²·s and RMSEs for significant wave height (SWH) and mean wave period (MWP) of 0.138 m and 1.331 s, respectively. At individual points, the spectral density bias is near zero, correlation coefficients range from 0.95 to 0.98, and the peak frequency RMSE is between 0.03 and 0.04 Hz. During a typical cold wave event, the model accurately reproduces the energy evolution and peak frequency shift. Buoy observations confirm that the model effectively tracks significant wave height trends under varying conditions. Moreover, applying a frequency-weighted loss function enhances the model’s ability to capture high-frequency spectral components, further improving prediction accuracy. Overall, the proposed method shows strong performance in spectrum prediction and provides a valuable approach for regional wave spectrum modeling. Full article

Frequency-hopping multiple access is widely adopted to blunt narrow-band jamming and limit spectral disclosure in cyber–physical systems, yet its practical resilience depends on three sequence-level properties. First, balancedness guarantees that every carrier is occupied equally often, removing spectral peaks that a jammer or energy detector could exploit. Second, a wide gap between successive hops forces any interferer to re-tune after corrupting at most one symbol, thereby containing error bursts. Third, a no-hit zone (NHZ) window with a zero pairwise Hamming correlation eliminates user collisions and self-interference when chip-level timing offsets fall inside the window. This work introduces an algebraic construction that meets the full set of requirements in a single framework. By threading a permutation over an integer ring and partitioning the period into congruent sub-blocks tied to the desired NHZ width, we generate balanced wide gap no-hit zone frequency-hopping (WG-NHZ FH) sequence sets. Analytical proofs show that (i) each sequence achieves the Lempel–Greenberger bound for auto-correlation, (ii) the family and zone sizes satisfy the Ye–Fan bound with equality, (iii) the hop-to-hop distance satisfies a provable WG condition, and (iv) balancedness holds exactly for every carrier frequency. Full article

To address the limitations of low detection efficiency and poor spatial resolution of traditional cable insulation diagnosis methods, a novel cable insulation diagnosis method based on impedance spectroscopy has been proposed. An impedance spectroscopy analysis model of the frequency response of high-voltage single-core cables under different aging conditions has been established. The initial classification of insulation condition is achieved based on the impedance phase deviation between the test cable and the reference cable. Under localized aging conditions, the impedance phase spectroscopy is more than twice as sensitive to dielectric changes as the amplitude spectroscopy. Leveraging this advantage, a multi-parameter diagnostic framework is developed that integrates key spectral features such as the first phase angle zero-crossing frequency, initial phase, and resonance peak amplitude. The proposed method enables quantitative estimation of aging severity, spatial extent, and location. This technique offers a non-invasive, high-resolution solution for advanced cable health diagnostics and provides a foundation for practical deployment of power system asset management. Full article

This paper presents a novel cochlear implant (CI) sound coding strategy called Spectral Feature Extraction (SFE). The SFE is a novel Fast Fourier Transform (FFT)-based Continuous Interleaved Sampling (CIS) strategy that provides less-smeared spectral cues to CI patients compared to Crystalis, a predecessor strategy used in Oticon Medical devices. The study also explores how the SFE can be enhanced into a Temporal Fine Structure (TFS)-based strategy named Spectral Event Extraction (SEE), combining spectral sharpness with temporal cues. Background/Objectives: Many CI recipients understand speech in quiet settings but struggle with music and complex environments, increasing cognitive effort. De-smearing the power spectrum and extracting spectral peak features can reduce this load. The SFE targets feature extraction from spectral peaks, while the SEE enhances TFS-based coding by tracking these features across frames. Methods: The SFE strategy extracts spectral peaks and models them with synthetic pure tone spectra characterized by instantaneous frequency, phase, energy, and peak resemblance. This deblurs input peaks by estimating their center frequency. In SEE, synthetic peaks are tracked across frames to yield reliable temporal cues (e.g., zero-crossings) aligned with stimulation pulses. Strategy characteristics are analyzed using electrodograms. Results: A flexible Frequency Allocation Map (FAM) can be applied to both SFE and SEE strategies without being limited by FFT bandwidth constraints. Electrodograms of Crystalis and SFE strategies showed that SFE reduces spectral blurring and provides detailed temporal information of harmonics in speech and music. Conclusions: SFE and SEE are expected to enhance speech understanding, lower listening effort, and improve temporal feature coding. These strategies could benefit CI users, especially in challenging acoustic environments. Full article

We present novel proofs of the Riemann hypothesis by extending the standard complex Riemann zeta function into a quaternionic algebraic framework. Utilizing λ-regularization, we construct a symmetrized form that ensures analytic continuation and restores critical-line reflection symmetry, a key structural property of the Riemann ξ(s) function. This formulation reveals that all nontrivial zeros of the zeta function must lie along the critical line Re(s) = 1/2, offering a constructive and algebraic resolution to this fundamental conjecture. Our method is built on convexity and symmetrical principles that generalize naturally to higher-dimensional hypercomplex spaces. We also explore the broader implications of this framework in quantum statistical physics. In particular, the λ-regularized quaternionic zeta function governs thermodynamic properties and phase transitions in Bose–Einstein condensates. This quaternionic extension of the zeta function encodes oscillatory behavior and introduces critical hypersurfaces that serve as higher-dimensional analogues of the classical critical line. By linking the spectral features of the zeta function to measurable physical phenomena, our work uncovers a profound connection between analytic number theory, hypercomplex geometry, and quantum field theory, suggesting a unified structure underlying prime distributions and quantum coherence. Full article

The transient operation of pump turbines generates significant flow-induced instabilities, prompting a comprehensive numerical investigation using the SST $k-\omega$ turbulence model to examine these instability effects throughout the complete operating range in turbine mode. This study specifically analyzes the evolutionary mechanisms of unsteady flow dynamics under ten characteristic off-design conditions while simultaneously characterizing the pressure fluctuation behavior within the vaneless space (VS). The results demonstrate that under both low-speed conditions and near-zero-discharge conditions, the VS and its adjacent flow domains exhibit pronounced flow instabilities with highly turbulent flow structures, while the pressure fluctuation amplitudes remain relatively small due to insufficient rotational speed or flow rate. Across the entire turbine operating range, the blade passing frequency (BPF) dominates the VS pressure fluctuation spectrum. Significant variations are observed in both low-frequency components (LFCs) and high-frequency, low-amplitude components (HF-LACs) with changing operating conditions. The HF-LACs exhibit relatively stable amplitudes but demonstrate significant variation in the frequency spectrum distribution across different operating conditions, with notably broader frequency dispersion under runaway conditions and adjacent operating points. The LFCs demonstrate significantly higher spectral density and amplitude magnitudes under high-speed, low-discharge operating conditions while exhibiting markedly reduced occurrence and diminished amplitudes in the low-speed, high-flow regime. This systematic investigation provides fundamental insights into the flow physics governing pump-turbine performance under off-design conditions while offering practical implications for optimizing transient operational control methodologies in hydroelectric energy storage systems. Full article

The scope of this study was to evaluate the response of Ce-doped gadolinium aluminum gallium garnet (GAGG:Ce) crystalline scintillator under medical X-ray irradiation for medical imaging applications. A 10 × 10 × 10 mm³ crystal was irradiated at X-ray tube voltages ranging from 50 kVp to 150 kVp. The crystal’s compatibility with several commercially available optical photon detectors was evaluated using the spectral matching factor (SMF) along with the absolute efficiency (AE) and the effective efficiency (EE). In addition, the energy-absorption efficiency (EAE), the quantum-detection efficiency (QDE) as well as the zero-frequency detective quantum detection efficiency DQE(0) were determined. The crystal demonstrated satisfactory AE values as high as 26.3 E.U. (where 1 E.U. = 1 μW∙m⁻²/(mR∙s⁻¹)) at 150 kVp, similar, or in some cases, even superior to other cerium-doped scintillator materials. It also exhibits adequate DQE(0) performance ranging from 0.99 to 0.95 across all the examined X-ray tube voltages. Moreover, it showed high spectral compatibility with commonly used photoreceptors in modern day such as complementary metal–oxide–semiconductors (CMOS) and charge-coupled-devices (CCD) with SMF values of 0.95 for CCD with broadband anti-reflection coating and 0.99 for hybrid CMOS blue. The aforementioned properties of this scintillator material were indicative of its superior efficiency in the examined medical energy range, compared to other commonly used scintillators. Full article

Low Earth orbit (LEO) satellites offer a revolutionary potential for positioning, navigation, and timing (PNT) services due to their stronger signal power and rapid geometric changes compared to traditional global navigation satellite systems (GNSS). However, dedicated LEO navigation systems face high costs, so opportunity navigation based on LEO satellites is a potential solution. This paper presents an orthogonal frequency division multiplexing (OFDM)-based LEO navigation system and analyzes its navigation performance. We use 5G new radio (NR) as the satellite transmitting signal and introduce the NR signal components that can be used for navigation services. The LEO NR system and a novel zero-padding correlation (ZPC) are introduced. This ZPC receiver can eliminate cyclic prefix (CP) and inter-carrier interference, thereby improving tracking accuracy. The power spectral density (PSD) for the NR navigation signal is derived, followed by a comprehensive analysis of tracking accuracy under different NR configurations (bandwidth, spectral allocation, and signal components). An extended Kalman filter (EKF) is proposed to fuse pseudorange and pseudorange rate measurements for real-time positioning. The simulations demonstrate an 80% improvement in ranging precision (3.0–4.5 cm) and 88.3% enhancement in positioning accuracy (5.61 cm) compared to conventional receivers. The proposed ZPC receiver can achieve centimeter-level navigation accuracy. This work comprehensively analyzes the navigation performance of the LEO NR system and provides a reference for LEO PNT design. Full article

The aim of this article is to critically and reliably assess the potential of current emotion recognition technologies for practical applications in human–computer interaction (HCI) systems. The study made use of two databases: one in English (RAVDESS) and another in Polish (EMO-BAJKA), both containing speech recordings expressing various emotions. The effectiveness of recognizing seven and eight different emotions was analyzed. A range of acoustic features, including energy features, mel-cepstral features, zero-crossing rate, fundamental frequency, and spectral features, were utilized to analyze the emotions in speech. Machine learning techniques such as convolutional neural networks (CNNs), long short-term memory (LSTM) networks, and support vector machines with a cubic kernel (cubic SVMs) were employed in the emotion classification task. The research findings indicated that the effective recognition of a broad spectrum of emotions in a subject-independent approach is limited. However, significantly better results were obtained in the classification of paired emotions, suggesting that emotion recognition technologies could be effectively used in specific applications where distinguishing between two particular emotional states is essential. To ensure a reliable and accurate assessment of the emotion recognition system, care was taken to divide the dataset in such a way that the training and testing data contained recordings of completely different individuals. The highest classification accuracies for pairs of emotions were achieved for Angry–Fearful (0.8), Angry–Happy (0.86), Angry–Neutral (1.0), Angry–Sad (1.0), Angry–Surprise (0.89), Disgust–Neutral (0.91), and Disgust–Sad (0.96) in the RAVDESS. In the EMO-BAJKA database, the highest classification accuracies for pairs of emotions were for Joy–Neutral (0.91), Surprise–Neutral (0.80), Surprise–Fear (0.91), and Neutral–Fear (0.91). Full article

A new deconvolution strategy, reverse curve fitting, was developed to determine peak positions and independent intensities of overlapping Fourier transform (FT) nuclear magnetic resonance (NMR) bands. From the third-order derivative of the overlapping band, the peak position was estimated from its zero-crossing point and the peak intensity was quantitated by partial curve matching with its primary maxima. Every matched peak in the overlapping band was dismembered in turn to weaken the overlap until an independent peak was filtered out. The deconvolution can be refined progressively by manually tuning the peak positions and peak widths. In a simulation study, a closely overlapped ¹³C NMR triplet (overlapping degrees between 0.5 and 1.0) at a signal-to-noise ratio (SNR) of 20:1 was quantitatively deconvoluted by our reverse curve fitting procedure with a routine denoising technique. The noise interference and denoising technique were also studied in the simulation. A real FT-NMR overlapping band of Ethylbenzene (300 MHz) was satisfactorily deconvoluted and compatible with higher resolution literature spectral data. A more complicated overlapping NMR band of Tetraphenyl porphyrin was studied as well. This new approach to the deconvolutions is applicable to other FT spectroscopies. Full article

To extend the spectral utilisation of In₂S₃, an In₂S₃/C₃N₄ nanocomposite was prepared. The effects of different sulphur sources, electrodes, and bias voltages on the optoelectronic performance were examined. Photoelectric properties in response to light sources with wavelengths of 405, 532, 650, 780, 808, 980, and 1064 nm were investigated using Au electrodes and the carbon electrodes with 5B pencil drawings. This study shows that the aggregation states of the In₂S₃/C₃N₄ nanocomposite possess photocurrent switching responses in the broadband region of the light spectrum. Combining two types of partially visible light-absorbing material extends utilisation to the near-infrared region. Impurities or defects embody an electron-donating effect. Since the energy levels of defects or impurities with an electron-donating effect are close to the conduction band, low-energy lights (especially NIR) can be utilised. The non-equilibrium carrier concentration (photogenerated electrons) of the nanocomposites increases significantly under NIR photoexcitation conditions. Thus, photoconductive behaviour is manifested. A good photoelectric signal was still measured when zero bias was applied. This demonstrates self-powered photoelectric response characteristics. Different sulphur sources significantly affect the photoelectric performance, suggesting that they create different defects that affect charge transport and base current noise. It is believed that interfacial interactions in the In₂S₃/C₃N₄ nanocomposite create a built-in electric field that enhances the separation and transfer of electrons and holes produced by light stimulation. The presence of the built-in electric field also leads to energy band bending, which facilitates the utilisation of the light with longer wavelengths. This study provides a reference for multidisciplinary applications. Full article