Search Results (73)

In recent years, deep learning-based end-to-end Text-To-Speech (TTS) models have made significant progress in enhancing speech naturalness and fluency. However, existing Variational Inference Text-to-Speech (VITS) models still face challenges such as insufficient pitch modeling, inadequate contextual dependency capture, and low inference efficiency in the decoder. To address these issues, this paper proposes an improved TTS framework named Q-VITS. Q-VITS incorporates Rotary Position Embedding (RoPE) into the text encoder to enhance long-sequence modeling, adopts a frame-level prior modeling strategy to optimize one-to-many mappings, and designs a style extractor based on a diffusion model for controllable style rendering. Additionally, the proposed decoder ConfoGAN integrates explicit F0 modeling, Pseudo-Quadrature Mirror Filter (PQMF) multi-band synthesis and Conformer structure. The experimental results demonstrate that Q-VITS outperforms the VITS in terms of speech quality, pitch accuracy, and inference efficiency in both subjective Mean Opinion Score (MOS) and objective Mel-Cepstral Distortion (MCD) and Root Mean Square Error (RMSE) evaluations on a single-speaker dataset, achieving performance close to ground-truth audio. These improvements provide an effective solution for efficient and controllable speech synthesis. Full article

The paper presents results of experimental research on tyre–road contact stress distributions, measured indoors for a motorsport slick tyre. The triaxial contact stress distributions have been measured using the complex transducer containing a transversal array of 30 sensing pins covering the entire contact patch width. Wheel displacement in the longitudinal direction was measured using a rotary encoder. The parameters allocated for the experimental programme have included different values of tyre inflation pressure, vertical load, camber angle and toe angle. All measurements were performed at low longitudinal speed in free-rolling conditions. The influence of tyre functional parameters on the contact patch shape and size has been discussed. The stress distributions on each orthogonal direction are presented in multiple formats, such as 2D graphs in which the curves show the stresses measured by each sensing element versus contact length; surfaces with stress values plotted as vertical coordinates versus contact patch length and width; and colour maps for stress distributions and orientations of shear stress vectors. The effects of different parameter types and values on stress distributions have been emphasised and analysed. Furthermore, the magnitude and position of local extreme values for each stress distribution have been investigated with respect to the above-mentioned tyre functional parameters. Full article

Background: High-intensity actions like accelerations and decelerations, often performed unilaterally, are crucial in elite female football but increase the risk of interlimb asymmetries and injury. Flywheel resistance training enhances eccentric strength, yet limited research has assessed how different inertial loads affect mechanical outputs in unilateral exercises. Purpose: This study investigated how two inertial loads (0.107 kg·m² and 0.133 kg·m²) influence power, acceleration, speed, and asymmetry during unilateral hip extensions in elite female footballers. Methods: Eighteen professional players (27 ± 4 years, 59.9 ± 6.5 kg, 168.2 ± 6.3 cm, BMI 21.2 ± 1.8) completed unilateral hip extensions on a conical flywheel under both inertia conditions. A rotary encoder measured peak/average power, acceleration, speed, and eccentric-to-concentric (E:C) ratios. Bilateral asymmetries between dominant (DL) and non-dominant (NDL) limbs were assessed. Paired t-tests and Cohen’s d were used for analysis. Results: Higher inertia reduced peak and mean acceleration and speed (p < 0.001, d > 0.8). Eccentric peak power significantly increased in the NDL (p < 0.001, d = 3.952). E:C ratios remained stable. Conclusions: Greater inertial loads reduce movement velocity but increase eccentric output in the NDL, offering potential strategies to manage neuromuscular asymmetries in elite female football players. Full article

The stability of gearbox bearings is crucial to the operational efficiency and safety of industrial equipment, as their faults can lead to downtime, economic losses, and safety risks. Traditional models face difficulties in handling complex industrial time-series data due to insufficient feature extraction capabilities and poor training stability. Although transformers show advantages in fault diagnosis, their ability to model local dependencies is limited. To improve feature extraction from time-series data and enhance model robustness, this paper proposes an innovative method based on the ViT. Time-series data were converted into two-dimensional images using polar coordinate transformation and Gramian matrices to enhance classification stability. A lightweight front-end encoder and depthwise feature extractor, combined with multi-scale depthwise separable convolution modules, were designed to enhance fine-grained features, while two-dimensional rotary position encoding preserved temporal information and captured temporal dependencies. The constructed RoPE-DWTrans model implemented a unified feature extraction process, significantly improving cross-dataset adaptability and model performance. Experimental results demonstrated that the RoPE-DWTrans model achieved excellent classification performance on the combined MCC5 and HUST gearbox datasets. In the fault category diagnosis task, classification accuracy reached 0.953, with precision at 0.959, recall at 0.973, and an F1 score of 0.961; in the fault category and severity diagnosis task, classification accuracy reached 0.923, with precision at 0.932, recall at 0.928, and an F1 score of 0.928. Compared with existing methods, the proposed model showed significant advantages in robustness and generalization ability, validating its effectiveness and application potential in industrial fault diagnosis. Full article

Rotator cuff tears are common in individuals over 40, and physical therapy is often prescribed post-surgery. However, access can be limited by cost, convenience, and insurance coverage. CuffLink is a telehealth rehabilitation system that integrates the Strengthening and Stabilization System mechanical exerciser with the interACTION mobile health platform. The system includes a triple-axis accelerometer (LSM6DSOX + LIS3MDL FeatherWing), a rotary encoder, a VL530X time-of-flight sensor, and two wearable BioMech Health IMUs to capture upper-limb motion. CuffLink is designed to facilitate controlled, home-based exercise while enabling clinicians to remotely monitor joint function. Concurrent validity and test–retest reliability were used to assess device accuracy and repeatability. The results showed moderate to good validity for shoulder rotation (ICC = 0.81), device rotation (ICC = 0.94), and linear tracking (from zero: ICC = 0.75 and RMSE = 2.41; from start: ICC = 0.88 and RMSE = 2.02) and good reliability (e.g., RMSEs as low as 1.66 cm), with greater consistency in linear tracking compared to angular measures. Shoulder rotation and abduction exhibited higher variability in both validity and reliability measures. Future improvements will focus on manufacturability, signal stability, and force sensing. CuffLink supports accessible, data-driven rehabilitation and holds promise for advancing digital health in orthopedic recovery. Full article

Addressing the significant challenge of speech enhancement in ultra-low-Signal-to-Noise-Ratio (SNR) scenarios for Unmanned Aerial Vehicle (UAV) voice communication, particularly under edge deployment constraints, this study proposes the Edge-Deployed Band-Split Rotary Position Encoding Transformer (Edge-BS-RoFormer), a novel, lightweight band-split rotary position encoding transformer. While existing deep learning methods face limitations in dynamic UAV noise suppression under such constraints, including insufficient harmonic modeling and high computational complexity, the proposed Edge-BS-RoFormer distinctively synergizes a band-split strategy for fine-grained spectral processing, a dual-dimension Rotary Position Encoding ($\mathrm{RoPE}$) mechanism for superior joint time–frequency modeling, and $\mathrm{FlashAttention}$ to optimize computational efficiency, pivotal for its lightweight nature and robust ultra-low-SNR performance. Experiments on our self-constructed DroneNoise-LibriMix (DN-LM) dataset demonstrate Edge-BS-RoFormer’s superiority. Under a −15 dB SNR, it achieves Scale-Invariant Signal-to-Distortion Ratio (SI-SDR) improvements of 2.2 dB over Deep Complex U-Net (DCUNet), 25.0 dB over the Dual-Path Transformer Network (DPTNet), and 2.3 dB over HTDemucs. Correspondingly, the Perceptual Evaluation of Speech Quality (PESQ) is enhanced by 0.11, 0.18, and 0.15, respectively. Crucially, its efficacy for edge deployment is substantiated by a minimal model storage of 8.534 MB, 11.617 GFLOPs (an 89.6% reduction vs. DCUNet), a runtime memory footprint of under 500MB, a Real-Time Factor (RTF) of 0.325 (latency: 330.830 ms), and a power consumption of 6.536 W on an NVIDIA Jetson AGX Xavier, fulfilling real-time processing demands. This study delivers a validated lightweight solution, exemplified by its minimal computational overhead and real-time edge inference capability, for effective speech enhancement in complex UAV acoustic scenarios, including dynamic noise conditions. Furthermore, the open-sourced dataset and model contribute to advancing research and establishing standardized evaluation frameworks in this domain. Full article

A novel temporally resolved settle-plate air sampler was developed using 3D printing technology to improve upon traditional passive air sampling methods. Conventional settle plates provide cumulative measurements of particle or microbial loads over an entire sampling period, lacking the temporal resolution necessary to identify specific contamination events. The described device integrates a petri plate within a 3D-printed housing featuring a narrow slit that exposes only a small portion of the plate to incoming particles. A rotary mechanism, driven by a mechanical clock motor, rotates the petri plate over 12 h, allowing for time-segmented sampling. Validation experiments demonstrated the device’s ability to accurately encode the temporal history of particle deposition using both aerosolized dyes and viable microbial spores. The device effectively correlated bioaerosol deposition with ambient wind conditions during outdoor sampling. The system is inexpensive (under USD 10), requires no specialized skills to assemble, and is compatible with existing settle plate methodologies. This innovation enhances the ability to conduct air quality assessments in critical environments, enabling data-driven decisions to mitigate contamination risks. Full article

A newly designed tilt-wing unmanned aerial vehicle (Tilt-wing UAV) requires a unified control strategy across rotary-wing, fixed-wing, and transition modes, introducing significant challenges. Existing control strategies typically rely on accurate modeling or extensive parameter tuning, which limits their adaptability to dynamically changing flight configurations. Although online reinforcement learning algorithms offer adaptability, they depend on real-world exploration, posing considerable safety and cost risks for safety-critical UAV applications. To address this challenge, we propose Temporal Sequence Constrained Q-learning (TSCQ), an offline RL framework that integrates an encoder–decoder with recurrent networks to capture temporal dependencies. The policy is further constrained within an offline dataset collected via hardware-in-the-loop simulation using a variational autoencoder, and a sequence-level prediction mechanism is introduced to ensure temporal consistency across action trajectories, thereby mitigating extrapolation error while preserving data fidelity. Experimental results demonstrate that TSCQ significantly outperforms gain scheduling, Model Predictive Control (MPC), and Batch-Constrained Q-learning (BCQ), reducing the RMSE of pitch angle by up to 53.3% and vertical velocity RMSE by approximately 33%. These findings underscore the potential of data-driven, safety-aware offline RL paradigms to enable robust and generalizable control strategies for tilt-wing UAVs. Full article

In this paper, we address the challenge of dynamic evolution of user preferences and propose an attribute-sequence-based recommendation model to improve the accuracy and interpretability of recommendation systems. Traditional approaches usually rely on item sequences to model user behavior, but ignore the potential value of attributes shared among different items for preference characterization. To this end, this paper innovatively replaces items in user interaction sequences with attributes, constructs attribute sequences to capture fine-grained preference changes, and reinforces the prioritization of current interests by maintaining the latest state of attributes. Meanwhile, the item–attribute relationship is modeled using LightGCN and a variant of GAT, fusing multi-level features using gated attention mechanism, and introducing rotary encoding to enhance the flexibility of sequence modeling. Experiments on four real datasets (Beauty, Video Games, Men, and Fashion) showed that the model in this paper significantly outperformed the benchmark model in both NDCG@10 and Hit Ratio@10 metrics, with a highest improvement of 6.435% and 3.613%, respectively. The ablation experiments further validated the key role of attribute aggregation and sequence modeling in capturing user preference dynamics. This work provides a new concept for recommender systems that balances fine-grained preference evolution with efficient sequence modeling. Full article

Named Entity Recognition (NER) is a fundamental task in Natural Language Processing (NLP) that supports applications such as information retrieval, sentiment analysis, and text summarization. While substantial progress has been made in NER for widely studied languages like English, Arabic presents unique challenges due to its morphological richness, orthographic ambiguity, and the frequent occurrence of nested and overlapping entities. This paper introduces a novel Arabic NER framework that addresses these complexities through architectural innovations. The proposed model incorporates a Hybrid Feature Fusion Layer, which integrates external lexical features using a cross-attention mechanism and a Gated Lexical Unit (GLU) to filter noise, while a Compound Span Representation Layer employs Rotary Positional Encoding (RoPE) and Bidirectional GRUs to enhance the detection of complex entity structures. Additionally, an Enhanced Multi-Label Classification Layer improves the disambiguation of overlapping spans and assigns multiple entity types where applicable. The model is evaluated on three benchmark datasets—ANERcorp, ACE 2005, and a custom biomedical dataset—achieving an F1-score of 93.0% on ANERcorp and 89.6% on ACE 2005, significantly outperforming state-of-the-art methods. A case study further highlights the model’s real-world applicability in handling compound and nested entities with high confidence. By establishing a new benchmark for Arabic NER, this work provides a robust foundation for advancing NLP research in morphologically rich languages. Full article

With the rapid increase in the number of Earth Observation Satellites (EOSs), research on autonomous mission scheduling has become increasingly critical for optimizing satellite sensor operations. While most existing studies focus on static environments or initial planning states, few address the challenge of dynamic request replanning for real-time sensor management. In this paper, we tackle the problem of multi-satellite rapid mission replanning under dynamic batch-arrival observation requests. The objective is to maximize overall observation revenue while minimizing disruptions to the original scheme. We propose a framework that integrates stochastic master-satellite mission allocation with single-satellite replanning, supported by reactive scheduling policies trained via deep reinforcement learning. Our approach leverages mission sequence modeling with attention mechanisms and time-attitude-aware rotary positional encoding to guide replanning. Additionally, scalable embeddings are employed to handle varying volumes of dynamic requests. The mission allocation phase efficiently generates assignment solutions using a pointer network, while the replanning phase introduces a hybrid action space for direct task insertion. Both phases are formulated as Markov Decision Processes (MDPs) and optimized using the PPO algorithm. Extensive simulations demonstrate that our method significantly outperforms state-of-the-art approaches, achieving a 15.27% higher request insertion revenue rate and a 3.05% improvement in overall mission revenue rate, while maintaining a 1.17% lower modification rate and achieving faster computational speeds. This demonstrates the effectiveness of our approach in real-world satellite sensor applications. Full article

Variable-speed electrical drive control typically relies upon a two-loop scheme, one for torque/speed and another for stator current control. In modern drive control methods, the actual mechanical speed is needed for both loops. In practical applications, the speed is often acquired by incremental rotary encoders. The most used method derives speed from an encoder pulse count during a fixed amount of time. It is known that this sensing method produces time delay in the speed feedback loop as well as fluctuations in the speed measurements. Time lags produce phase loss that has potentially negative effects on the overall drive performance. Nevertheless, the pulse counting method is favored in most cases due to its simplicity and existing support for its use in digital signal processors. In this paper, a new speed sensing method is proposed to reduce time lag without incurring increased fluctuations. The proposal uses a novel transient detector to determine the actual operational regime of the drive: transient or stationary. Transient detection is not based on measured speeds but works directly with the train of incoming encoder pulses. The method is designed to work well with established digital signal processor routines. The proposal is assessed through experimentation on a real five-phase induction motor. Full article

This study aimed to validate the effectiveness of the Articulation Motion Assessment System (AMAS), a joint kinematic evaluation system, for clinical applications. AMAS enables synchronised measurement using neurophysiological indicators, overcoming laboratory setting limitations. We compared AMAS-based ankle joint kinematic evaluations, particularly the sagittal and frontal plane angles, with two-dimensional (2D) motion analysis to determine the validity and reliability of AMAS. Both AMAS and 2D motion analysis reliably detected significant differences in angles within the sagittal and frontal planes. Correlation analysis revealed a significant moderate-to-strong correlation between the AMAS and the conventional method of 2D motion analysis, proving the measurement validity of the AMAS (ρ = 0.53–0.77 for sagittal plane angles; ρ = 0.46–0.72 for frontal plane angles). The average root mean squared error (RMSE) was significantly lower in AMAS (10.90 ± 2.93° for sagittal plane angles; 13.44 ± 1.09° for frontal plane angles) than in the inertial sensor-based three-dimensional (3D) motion analysis. Reliability analysis revealed high reliability of measurements (intraclass correlation coefficients (ICC) ≥ 0.76). However, the Bland–Altman analysis identified a slightly lower fixed bias, which was observed as a characteristic of each measurement system. The AMAS accurately detects ankle joint angles without being constrained by measurement environment limitations. Synchronised measurements using neurophysiological indicators potentially contribute to understanding ankle joint control mechanisms and developing rehabilitation strategies. Full article

To address the issue of efficiently reusing the massive amount of unstructured knowledge generated during the handling of track circuit equipment faults and to automate the construction of knowledge graphs in the railway maintenance domain, it is crucial to leverage knowledge extraction techniques to efficiently extract relational triplets from fault maintenance text data. Given the current lag in joint extraction technology within the railway domain and the inefficiency in resource utilization, this paper proposes a joint extraction model for track circuit entities and relations, integrating Global Pointer and tensor learning. Taking into account the associative characteristics of semantic relations, the nesting of domain-specific terms in the railway sector, and semantic diversity, this research views the relation extraction task as a tensor learning process and the entity recognition task as a span-based Global Pointer search process. First, a multi-layer dilate gated convolutional neural network with residual connections is used to extract key features and fuse the weighted information from the 12 different semantic layers of the RoBERTa-wwm-ext model, fully exploiting the performance of each encoding layer. Next, the Tucker decomposition method is utilized to capture the semantic correlations between relations, and an Efficient Global Pointer is employed to globally predict the start and end positions of subject and object entities, incorporating relative position information through rotary position embedding (RoPE). Finally, comparative experiments with existing mainstream joint extraction models were conducted, and the proposed model’s excellent performance was validated on the English public datasets NYT and WebNLG, the Chinese public dataset DuIE, and a private track circuit dataset. The F1 scores on the NYT, WebNLG, and DuIE public datasets reached 92.1%, 92.7%, and 78.2%, respectively. Full article

Introduction: This study aimed to evaluate the validity of the HandyGym portable flywheel device with an integrated rotary encoder in measuring force and power during iso-inertial exercises compared to a traditional reference system. Methods: In total, 10 trained volunteers (3 women, 7 men; age 25.2 ± 3.8 years) performed half-squats with five different load configurations using the HandyGym device. Concurrent measurements were obtained from HandyGym’s rotary encoder and a criterion system (MuscleLab 6000 strain gauge and linear encoder). Five load configurations were tested, with 15 repetitions recorded per condition. The validity of the HandyGym measurements was assessed through mean bias, typical error of estimation (TEE), and Pearson correlation coefficients, with Bland–Altman plots used to analyze the agreement between the two systems. Results: The HandyGym showed high correlations with the reference system for both force (r = 0.76–0.90) and power (r = 0.60–0.94). However, systematic biases were observed, with the HandyGym consistently underestimating force and power at lower loads and overestimating power at higher loads. The TEE values indicated moderate to large errors, particularly in power measurements. Conclusion: The HandyGym provides valid force measurements with moderate bias, suitable for general monitoring. However, power measurements are less consistent, especially at higher loads, limiting the device’s utility for precise assessments. Adjustments or corrections may be necessary for accurate application in professional contexts. Full article