检索结果-内蒙古大学图书馆

Elevating Robust Multi-Talker ASR by Decoupling Speaker Separation and Speech Recognition

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Yang, Yufeng Taherian, Hassan Kalkhorani, Vahid Ahmadi Wang, DeLiang Department of Computer Science and Engineering The Ohio State University 2015 Neil Avenue ColumbusOH43210 United States Center for Cognitive and Brain Sciences The Ohio State University 1835 Neil Ave ColumbusOH43210 United States

Despite the tremendous success of automatic speech recognition (ASR) with the introduction of deep learning, its performance is still unsatisfactory in many real-world multi-talker scenarios. Speaker separation excels in separating individual talkers but, as a frontend, it introduces processing artifacts that degrade the ASR backend trained on clean speech. As a result, mainstream robust ASR systems train the backend on noisy speech to avoid processing artifacts. In this work, we propose to decouple the training of the speaker separation frontend and the ASR backend, with the latter trained on clean speech only. Our decoupled system achieves 5.1% word error rates (WER) on the Libri2Mix dev/test sets, significantly outperforming other multi-talker ASR baselines. Its effectiveness is also demonstrated with the state-of-the-art 7.60%/5.74% WERs on 1-ch and 6-ch SMS-WSJ. Furthermore, on recorded LibriCSS, we achieve the speaker-attributed WER of 2.92%. These state-of-the-art results suggest that decoupling speaker separation and recognition is an effective approach to elevate robust multi-talker ASR. Copyright © 2025, The Authors. All rights reserved.

关键词： Speech recognition

Contract-Inspired Contest Theory for Controllable Image Generation in Mobile Edge Metaverse

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Liu, Guangyuan Du, Hongyang Wang, Jiacheng Niyato, Dusit Kim, Dong In College of Computing and Data Science The Energy Research Institute @ NTU Interdisciplinary Graduate Program Nanyang Technological University Singapore Department of Electrical and Electronic Engineering University of Hong Kong Hong Kong College of Computing and Data Science Nanyang Technological University Singapore Department of Electrical and Computer Engineering Sungkyunkwan University Korea Republic of

The rapid advancement of immersive technologies has propelled the development of the Metaverse, where the convergence of virtual and physical realities necessitates the generation of high-quality, photorealistic images to enhance user experience. However, generating these images, especially through Generative Diffusion Models (GDMs), in mobile edge computing environments presents significant challenges due to the limited computing resources of edge devices and the dynamic nature of wireless networks. This paper proposes a novel framework that integrates contract-inspired contest theory, Deep Reinforcement Learning (DRL), and GDMs to optimize image generation in these resource-constrained environments. The framework addresses the critical challenges of resource allocation and semantic data transmission quality by incentivizing edge devices to efficiently transmit high-quality semantic data, which is essential for creating realistic and immersive images. The use of contest and contract theory ensures that edge devices are motivated to allocate resources effectively, while DRL dynamically adjusts to network conditions, optimizing the overall image generation process. Experimental results demonstrate that the proposed approach not only improves the quality of generated images but also achieves superior convergence speed and stability compared to traditional methods. This makes the framework particularly effective for optimizing complex resource allocation tasks in mobile edge Metaverse applications, offering enhanced performance and efficiency in creating immersive virtual environments. © 2025, CC BY.

关键词： Mobile edge computing

A Secure NFC Cardless Cash Withdrawal System 5th

学校读者我要写书评

暂无评论

A Secure NFC Cardless Cash Withdrawal System

5th International Conference on Financial Technology, ICFT 2024

作者： Hwang, Min-Shiang Lin, Cheng-Ying Chang, Chun-Hsien Yang, Cheng-Ying Department of Computer Science and Information Engineering Asia University Taichung Taiwan Fintech and Blockchain Research Center Asia University Taichung Taiwan The Ph.D. Program in Artificial intelligence Asia University Taichung Taiwan Department of Management Information Systems National Chung Hsing University Taichung Taiwan Department of Computer Science University of Taipei Taipei Taiwan

ISBN: (纸本)9789819638109

As IoT technology grows, several different payment methods are available now. NFC (Near Field Communication) as a short-range communication technology has been applied in many fields, such as contactless payment and electronic tickets. ATM (Auto Teller Machine) also benefits from NFC technology as one way of cardless cash withdrawal. However, there are many security problems, and convenience problems exist. In security, certain kinds of attacks, such as man-in-the-middle, replay, and brute-force attacks, must be considered. These attacks may influence user privacy. Furthermore, considering user experience, existing methods may not make users feel enough convenience when using. To enhance security and user experience, we proposed a secure NFC cardless cash withdrawal system using time stamps and nonce to reduce the risk of attack and design more effortless operations for users to enhance user experience. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Automatic teller machines

Speech Separation for Low-Resource Languages

学校读者我要写书评

暂无评论

Speech Separation for Low-Resource Languages

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Marvin Borsdorf Zexu Pan Pascal Himmelmann Haizhou Li Tanja Schultz Machine Listening Lab University of Bremen Bremen Germany Alibaba Group Singapore Department of Mathematics and Computer Science FernUniversität in Hagen Hagen Germany School of Data Science Shenzhen Research Institute of Big Data The Chinese University of Hong Kong Shenzhen China Department of Electrical and Computer Engineering National University of Singapore Singapore Cognitive Systems Lab University of Bremen Bremen Germany

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Speech separation aims to equip machines with the human ability of selective listening, i.e. to focus attention on specific information in spoken communication. Studies have shown that the language spoken in a cocktail party scenario matters. While the development of speech separation models can leverage extensive databases, for the majority of languages only very limited data is available. This work presents the very first study on speech separation for low-resource languages. We choose blind source separation as the task to be studied and analyze three strategies to overcome the data scarcity of two low-resource languages from the GlobalPhoneMS2 database. We show that data from other languages can be used to develop models that work for low-resource languages. Finetuning additionally boosts the performance, and training on multiple languages increases both performance and robustness. We show that dynamic mixing in the development helps to find a trade-off between performance and development time.

关键词： Training Measurement Databases Signal processing Data models Blind source separation Robustness Acoustics Speech processing

2D Polynomial Regression Modeling for Predictive Analysis of Glucose Concentration in NP-Model Fractal Microwave Sensor Using Frequency and Transmission Coefficient Disparities

学校读者我要写书评

暂无评论

2D Polynomial Regression Modeling for Predictive Analysis of...

International Electrical engineering Congress (iEECON)

作者： Nutyada Nantapanich Pisit Mahatthanajatuphat Norakamon Wongsin Wilaiporn Lee Department of Electronics and Telecommunication Engineering Faculty of Engineering Rajamangala University of Technology Thanyaburi Pathum Thani Thailand Regular Program in Mathematics-Science Yothinburana School Bangkok Thailand Department of Electrical and Computer Engineering Faculty of Engineering King Mongkut's University of Technology North Bangkok Bangkok Thailand

ISBN: (数字)9798331543952

ISBN: (纸本)9798331543969

The article introduces a two-dimensional polynomial regression model for the predictive analysis of glucose concentration in a fractal microwave sensor NP model, utilizing frequency and transmission coefficient differences to examine their correlation with glucose concentration levels. The sensor employed in this investigation is fabricated on an economical FR-4 substrate to minimize costs. The sensor structure is developed with the fractal structure reduction method established by the NP model to operate at a frequency of roughly 2.97 GHz. The CST Studio software is utilized to analyze the findings from the simulation of the suggested sensor structure. Furthermore, the proposed sensor is integrated into a test apparatus comprising a sensor platform and a liquid tube constructed from PPA material. The liquid under examination is a glucose and water mixture at varying concentrations. It was analyzed using a VNA instrument to evaluate its frequency response and the magnitude of the alteration in the transmission coefficient. The values are utilized to create a two-dimensional polynomial regression model. The experiments demonstrated that the proposed sensor model possesses an Adjusted R² value of 0.99499 and an average error of approximately 3.57% for glucose concentration analysis throughout the range of 5% to 30%, with sensitivities of 0.00879 dB/% for insertion loss and 0.092 MHz/% for frequency variation, facilitating accurate glucose concentration prediction analysis.

关键词： Analytical models Adaptation models Correlation Liquids Predictive models Microwave sensors Polynomials Fractals Glucose Predictive analytics

Joint Source Decomposition of Concurrent EEG-FMRI Data in Epilepsy and Control Groups

学校读者我要写书评

暂无评论

Joint Source Decomposition of Concurrent EEG-FMRI Data in Ep...

IEEE International Symposium on Biomedical Imaging

作者： J. Mason Harding Haatef Pourmotabbed Yamin Li Kimberly Rogge-Obando Kate Wang Sarah E. Goodale Shiyu Wang Camden Bibro Bergen Allee Caroline Martin Victoria L. Morgan Dario J. Englot Catie Chang Department of Electrical and Computer Engineering Vanderbilt University Department of Biomedical Engineering Vanderbilt University Department of Computer Science Vanderbilt University Neuroscience Graduate Program Vanderbilt University Vanderbilt Memory and Alzheimer's Center Vanderbilt University Medical Center Department of Neurological Surgery Vanderbilt University Medical Center Department of Radiology Vanderbilt University Medical Center

ISBN: (数字)9798331520526

ISBN: (纸本)9798331520533

EEG and fMRI are complementary, noninvasive technologies for investigating human brain function. These modalities have been used to uncover large-scale functional networks and their disruptions in clinical populations. Given the high temporal resolution of EEG and high spatial resolution of fMRI, integrating these modalities can provide a more holistic understanding of brain activity. This work explores a multimodal source decomposition technique for extracting shared modes of temporal variation between fMRI BOLD signals and EEG spectral power fluctuations in the resting state. The resulting components are then compared between patients with focal epilepsy and controls, revealing multimodal network differences between groups.

关键词： Neuroimaging Brain Fluctuations Epilepsy Functional magnetic resonance imaging Electroencephalography Pattern recognition Hemodynamics Spatial resolution Signal resolution

Effective Dimension Aware Fractional-Order Stochastic Gradient Descent for Convex Optimization Problems

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Partohaghighi, Mohammad Marcia, Roummel Chen, YangQuan Electrical Engineering and Computer Science graduate program University of California Merced United States Department of Applied Mathematics University of California Merced MercedCA United States Lab Department of Mechanical Engineering School of Engineering University of California MercedCA95343 United States

Fractional-order stochastic gradient descent (FOSGD) leverages a fractional exponent to capture long-memory effects in optimization, yet its practical impact is often constrained by the difficulty of tuning and stabilizing this exponent. In this work, we introduce 2SED Fractional-Order Stochastic Gradient Descent (2SEDFOSGD), a novel method that synergistically combines the Two-Scale Effective Dimension (2SED) algorithm with FOSGD to automatically calibrate the fractional exponent in a data-driven manner. By continuously gauging model sensitivity and effective dimensionality, 2SED dynamically adjusts the exponent to curb erratic oscillations and enhance convergence rates. Theoretically, we demonstrate how this dimension-aware adaptation retains the benefits of fractional memory while averting the sluggish or unstable behaviors frequently observed in naive fractional SGD. Empirical evaluations across multiple benchmarks confirm that our 2SED-driven fractional exponent approach not only converges faster but also achieves more robust final performance, suggesting broad applicability for fractional-order methodologies in large-scale machine learning and related domains. © 2025, CC BY.

关键词： Convex optimization

Unbiased Parameter Estimation for Bayesian Inverse Problems

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Chada, Neil K. Jasra, Ajay Maama, Mohamed Tempone, Raul Department of Mathematics City University of Hong Kong China School of Data Science The Chinese University of Hong Kong Shenzhen CN Shenzhen China Applied Mathematics and Computational Science Program Computer Electrical and Mathematical Sciences and Engineering Division King Abdullah University of Science and Technology Thuwal23955-6900 Saudi Arabia

In this paper we consider the estimation of unknown parameters in Bayesian inverse problems. In most cases of practical interest, there are several barriers to performing such estimation, This includes a numerical approximation of a solution of a differential equation and, even if exact solutions are available, an analytical intractability of the marginal likelihood and its associated gradient, which is used for parameter estimation. The focus of this article is to deliver unbiased estimates of the unknown parameters, that is, stochastic estimators that, in expectation, are equal to the maximizer of the marginal likelihood, and possess no numerical approximation error. Based upon the ideas of [4] we develop a new approach for unbiased parameter estimation for Bayesian inverse problems. We prove unbiasedness and establish numerically that the associated estimation procedure is faster than the current state-of-the-art methodology for this problem. We demonstrate the performance of our methodology on a range of problems which include a PDE and ODE. Copyright © 2025, The Authors. All rights reserved.

关键词： Markov processes

Exploring Augmented Reality's Influence on cognitive Load and Emotional Dynamics within UAV Training Environments

学校读者我要写书评

暂无评论

IEEE Transactions on Learning Technologies 2025年

作者： Rahimi, Fatema Sadeghi-Niaraki, Abolghasem Song, Houbing Wang, Huihui Choi, Soo-Mi Sejong University Department of Computer Science and Engineering Convergence Engineering for Intelligent Drone XR Research Center Seoul Korea Republic of Department of Information Systems BaltimoreMD21250 United States St. Bonaventure University Cybersecurity Program St. BonaventureNY14778 United States

This study investigates the cognitive and emotional processes involved in augmented reality (AR)-based learning. The study looks at learning outcomes, emotional responses, meditation, and attention using a comprehensive approach that includes self-assessment, EEG data gathering, and post-experiment questionnaires. Twelve participants, selected based on their English proficiency and lack of prior knowledge of the course material, engaged in AR-based learning, while a baseline reading condition was included to contextualize cognitive and emotional engagement. The study findings indicate that the AR group's participants demonstrated notably elevated attention and meditation levels, indicating heightened engagement and focus that is advantageous for efficient assimilation and retention of knowledge. Furthermore, AR learners reported feeling less tired and exhausted, which may have mitigated the negative emotional states that are frequently connected to learning activities. However, no significant differences in negative emotions were observed between the reading and AR groups. These results emphasize the value of customized AR environments for education goals and the need for more study to maximize learning outcomes and affective experiences in AR learning contexts. © 2008-2011 IEEE.

关键词： engineering education