检索结果-内蒙古大学图书馆

International Symposium on Chinese Spoken language Processing

作者： Ao Zhang Fan Yu Kaixun Huang Lei Xie Longbiao Wang Eng Siong Chng Hui Bu Binbin Zhang Wei Chen Xin Xu Audio Speech and Language Processing Group (ASLP@NPU) School of Computer Science Northwestern Polytechnical University Xi’an China Tianjin Key Laboratory of Cognitive Computing and Application College of Intelligence and Computing Tianjin University Tianjin China School of Computer Science and Engineering Nanyang Technological University Singapore Beijing Shell Shell Technology Co. Ltd Beijing China WeNet Open Source Community Li Auto Inc. Beijing China

ISBN: (纸本)9798350397970

This paper summarizes the outcomes from the ISCSLP 2022 Intelligent Cockpit Speech Recognition Challenge (ICSRC). We first address the necessity of the challenge and then introduce the associated dataset collected from a new-energy vehicle (NEV) covering a variety of cockpit acoustic conditions and linguistic contents. We then describe the track arrangement and the baseline system. Specifically, we set up two tracks in terms of allowed model/system size to investigate resource-constrained and -unconstrained setups, targeting to vehicle embedded as well as cloud ASR systems respectively. Finally we summarize the challenge results and provide the major observations from the submitted systems.

关键词： Measurement Industries Visualization Target tracking Speech recognition Linguistics Acoustics

来源：评论

学校读者我要写书评

暂无评论

The Algorithm of Voiceprint Recognition Model based DNN-RELIANCE

The Algorithm of Voiceprint Recognition Model based DNN-RELI...

引用

International Conference on Computer engineering and Applications (ICCEA)

作者： Jing Zhang School of Information Science and Technology Laboratory of Language Engineering and Computing Guangdong University of Foreign Studies Guangzhou City China

ISBN: (数字)9781728159041

ISBN: (纸本)9781728159058

The traditional voiceprint recognition model (Gaussian hybrid model, hidden Markov model) is very dependent on the scale of speaker speech. Higher recognition performance often requires a large number of trained speaker speech to achieve, and has such shortcomings as environmental noise sensitivity and long training time, besides, it is difficult in convergence, the application of voiceprint recognition in practice is limited accordingly. The deep neural network has the ability of autonomous learning, and it can extract the deep speaker speech features that are not sensitive to environmental noise according to the target, and the extracted speaker speech features can be well classified and identified with the powerful pattern classification ability. So the deep neural network (DNN) is used as the voiceprint recognition model in this paper. However, the experiment shows that the DNN-based voiceprint recognition system still has a low accuracy rate of rejection of counterfeiters. For this reason, the trust degree and label distance are introduced, and the two-order judgment structure based on DNN-RLIANCE algorithm is proposed. Experiments show that the algorithm has short training and recognition time but high recognition accuracy, especially the system's correct rejection rate for impostors is improved.

关键词： Neural networks Performance analysis Real-time systems

来源：评论

学校读者我要写书评

暂无评论

CESMP: Chinese-English Segment-aligned Multi-field Patent Data

CESMP: Chinese-English Segment-aligned Multi-field Patent Da...

引用

IEEE International Conference on Cloud computing and Intelligence Systems (CCIS)

作者： Wuying Liu Lin Wang Fumao Hu Laboratory of Language Engineering and Computing Guangdong University of Foreign Studies Guangzhou China Xianda College of Economics and Humanities Shanghai International Studies University Shanghai China School of Foreign Languages Luoyang Institute of Science and Technology Luoyang China

ISBN: (纸本)9781665441506

Patent data from various countries in the world implies the essence of scientific discovery and technological innovation of all human beings, but language differences have become a huge obstacle to patent data retrieval and communication. We hope to build a bridge from Chinese to English in the patent domain, so that English speakers can make better use of Chinese patent data. With the help of natural language processing technologies such as optical character recognition, Chinese text processing, machine translation and English text processing, we construct digital Chinese-English segment-aligned multi-field patent (CESMP) data from scanned Chinese patents. The current CESMP data consists of 610,310 patent documents in XML format. Each patent document contains six required fields (date, publication, ipc, title, abstract, and claim) and four optional fields (cpc, wipo, originalapplicant, and currentowner), among which the wipo, title, abstract, and claim fields are aligned with Chinese and English segments. Supported by well-structured bilingual patent data, on the one hand, the resource construction algorithms can efficiently build a bilingual patent dictionary and a parallel patent segment bank; on the other hand, the deep natural language processing algorithms can be effectively implemented into many practical intelligent applications such as cross-language patent retrieval, patent spam filtering, patent network analysis, patent machine translation, etc.

关键词： Optical filters Patents Technological innovation XML Network analyzers Adaptive optics Optical character recognition software

来源：评论

学校读者我要写书评

暂无评论

Improving Naturalness and Controllability of Sequence-to-Sequence Speech Synthesis by Learning Local Prosody Representations

Improving Naturalness and Controllability of Sequence-to-Seq...

引用

IEEE International Conference on Acoustics, Speech and Signal Processing

作者： Cheng Gong Longbiao Wang Zhenhua Ling Shaotong Guo Ju Zhang Jianwu Dang Tianjin Key Laboratory of Cognitive Computing and Application College of Intelligence and Computing Tianjin University Tianjin P.R.China National Engineering Laboratory for Speech and Language Information Processing University of Science and Technology of China Hefei P.R.China Huiyan Technology (Tianjin) Co. Ltd Japan Advanced Institute of Science and Technology Ishikawa Japan

ISBN: (纸本)9781728176055;9781728176062

State-of-the-art neural text-to-speech (TTS) networks are trained with a large amount of speech data, which significantly improves the quality of synthetic speech compared with traditional approaches. However, the prosody and controllability of the generated speech is still insufficient, especially in tonal languages. Moreover, the generated prosody is solely defined by the input text, which does not allow for different styles for the same sentence or words. In this study, we extended Tacotron2 with a pitch prediction task to capture discrete pitch-related representations. Specifically, the learned pitch-related suprasegmental information is fed simultaneously with traditional character features into the decoder to generate final Mel spectrogram. Experiments show that the proposed method can improve the quality of the generated speech (mean opinion score of 4.37 vs. 4.22). Moreover, we demonstrated that we can easily achieve word-level pitch control during generation by changing local pitch-related representations before passing them to the decoder network.

关键词： Conferences Neural networks Controllability Acoustics Decoding Speech synthesis Task analysis

来源：评论

学校读者我要写书评

暂无评论

An Effective Learning Management System for Revealing Student Performance Attributes

arXiv

引用

arXiv 2024年

作者： Zhang, Xinyu Lee, Vincent C.S. Xu, Duo Chen, Jun Obaidat, Mohammad S. The School of Electronics and Information Northwestern Polytechnical University Xi’an710072 China The Department of Data Science and Artificial Intelligence Faculty of IT Monash University Melbourne3800 Australia Guangyuan Foreign Language School Guangyuan628018 China Chongqing Institute for Brain and Intelligence Guangyang Bay Laboratory Chongqing400064 China The King Abdullah II School of Information Technology The University of Jordan Amman11942 Jordan School of Computer and Communication Engineering University of Science and Technology Beijing Beijing100083 China Department of Computational Intelligence School of Computing SRM University SRM Nagar TN Kattankulathur603203 India School of Engineering The Amity University UP Noida201301 India

Contribution: A learning management system (LMS) incorporated with an advanced educational data mining module is proposed, as a means to mine efficiently from student performance records to provide valuable insights for educators in helping plan effective learning pedagogies, improve curriculum design, and guarantee quality of teaching. Background: An LMS streamlines the management of the teaching process in a centralized place, recording, tracking, and reporting the delivery of educational courses and student performance. Educational knowledge discovery from such an elearning system plays a crucial role in rule regulation, policy establishment, and system development. However, existing LMSs do not have embedded mining modules to directly extract knowledge. As educational modes become more complex, educational data mining efficiency from those heterogeneous student learning behaviours is gradually degraded. Intended outcomes: The design and application of the LMS enable educators to learn from past student performance experiences, empowering them to guide and intervene with students in time, and eventually improve their academic success. Application design: This study proposes an effective LMS which processes the stored data through an advanced educational data mining module. The mining module utilizes probability-based frequent pattern mining for generating candidate itemsets and updating LMS databases in each iteration for complexity reduction. In addition, the conditional probability increment ratio (CPIR) is deployed as an exceptionality measure to mine effective common and exception rules simultaneously. The iteratively generated rules can reveal academic performance patterns so that have merits in learning personal goal achievement. Findings: Through two illustrative case studies, experimental results demonstrate increased mining efficiency of the proposed mining module without information loss compared to classic educational mining algorithms. The mined knowl

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

The Design and Construction of the Corpus of China English

The Design and Construction of the Corpus of China English

引用

作者： Lixin Xia Yun Xia Laboratory of Language Engineering and Computing of Guangdong University of Foreign Studies Nanfang College of Sun Yat-Sen University/Guangdong University of Foreign Studies

The paper describes the design and construction of the Corpus of China English(CCE). With the emergence of China English as a developing variety in the family of the world Englishes, more and more research has been done to explore its use in China. In order to provide a reliable resource for researchers in the field of China English, CCE was built with due consideration given to its representativeness and authenticity. The general principles for the corpus were authentic, representative and manageable in *** was composed of more than 13,962,102 tokens in 15,333 texts evenly divided between the following four genres: newspapers,magazines, fiction and academic writings. The texts cover a wide range of domains, such as news, financial, politics, environment,social, culture, technology, sports, education, philosophy, literary,etc. It will be a helpful resource for research on China English,computational linguistics, natural language processing and corpus linguistics. Moreover, it is also a valuable resource for English language teaching in the context of English as a lingua franca(ELF)in which the theories and practice of English language teaching shift from teaching the "Standard" English to English as a lingua franca. With the corpus evidence, English teaching textbooks and materials can be developed with due consideration given to China English.

关键词： Corpus China English Computational linguistics Corpus linguistics Natural language Processing English language teaching

来源：评论

学校读者我要写书评

暂无评论

Dynamic Prefix-Tuning for Generative Template-based Event Extraction

arXiv

引用

arXiv 2022年

作者： Liu, Xiao Huang, Heyan Shi, Ge Wang, Bo School of Computer Science and Technology Beijing Institute of Technology China Beijing Engineering Research Center of High Volume Language Information Processing and Cloud Computing Applications China Key Laboratory of Intelligent Information Processing and Information Security Ministry of Industry and Information Technology China Southeast Academy of Information Technology Beijing Institute of Technology China Faculty of Information Technology Beijing University of Technology China

We consider event extraction in a generative manner with template-based conditional generation. Although there is a rising trend of casting the task of event extraction as a sequence generation problem with prompts, these generation-based methods have two significant challenges, including using suboptimal prompts and static event type information. In this paper, we propose a generative template-based event extraction method with dynamic prefix (GTEE-DYNPREF) by integrating context information with type-specific prefixes to learn a context-specific prefix for each context. Experimental results show that our model achieves competitive results with the state-of-the-art classification-based model ONEIE on ACE 2005 and achieves the best performances on ERE. Additionally, our model is proven to be portable to new types of events effectively. Copyright © 2022, The Authors. All rights reserved.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

A three-way utility decision model with multi-label based on fuzzy neighborhood environments

引用

Applied Soft computing 2025年 180卷

作者： Qi, Zhaohui Hu, Jiafeng Li, Hui Dai, Jianhua Hunan Provincial Laboratory of Intelligent Computing and Language Information Processing Hunan Normal University Changsha Hunan 410081 China College of Information Science and Engineering Hunan Normal University Changsha 410081 China Institute of Interdisciplinary Studies Hunan Normal University Changsha 410081 China

Recently, there has been extensive research conducted on the Multi-Attribute Decision Making (MADM) problem, particularly focusing on the growing number of attributes involved. The importance of attribute weight has become a crucial factor in MADM scenarios. Conventional approaches for determining attribute weights are no longer sufficient for this domain. The primary obstacle is to efficiently assign attribute weights in order to get effective decision-making. This research presents a novel three-way utility decision (3WUD) model that combines multi-label and fuzzy neighborhood environments. Initially, we construct a Multi-Label Fuzzy Neighborhood Decision System (MLFNDS) utilizing an innovative method to calculate the multi-label conditional probability. Afterward, we determine the label importance and create a utility function that takes into account multi-label. Furthermore, we present a novel approach to assess the attribute importance by considering both algebraic and information-theoretic viewpoints. Ultimately, we have constructed a 3WUD model within multi-label fuzzy neighborhood environments to determine the definitive ranking of all objects. Additionally, four evaluation indicators are utilized to assess the efficiency, logic, and excellence of the suggested methodology. In summary, this paper provides a new perspective on three-way decision (3WD) and MADM research and expands its application scope. In addition, it also provides a new method and perspective for determining attribute weights in MADM. © 2025

关键词： Multi-label conditional probability Multi-label decision making Multi-label fuzzy neighborhood decision system Multi-label utility function Three-way decision

来源：评论

学校读者我要写书评

暂无评论

Closed-Loop Safe Correction for Reinforcement Learning Policy 1

引用

4th International Conference on Ubiquitous Security, UbiSec 2024

作者： Yi, Zhi Lv, Qi Chen, Shuhong Liang, Ying Dai, Yinglong Hunan Provincial Key Laboratory of Intelligent Computing and Language Information Processing College of Information Science and Engineering Hunan Normal University Changsha410081 China School of Computer Science and Cyber Engineering Guangzhou University Guangzhou510006 China Blockchain Innovation Lab Swinburne University of Technology MelbourneVIC3122 Australia Molecular Nutrition Branch National Engineering Research Center of Rice and By-product Deep Processing College of Food Science and Engineering Central South University of Forestry and Technology Hunan Changsha410004 China

ISBN: (数字)9789819648368

ISBN: (纸本)9789819648351

Trial and error learning is an approach with uncertain consequences. How to maintain policy security, stability, and efficiency under controlled circumstances, posing a significant academic challenge. Such as Reinforcement Learning (RL) leverages the neural network parameter space to identify the optimal solution for achieving the optimal mapping of the policy space. However, even a slight fluctuation in the parameter space can lead to policy collapse. To solve it, we propose a method called Closed-loop Safe Correction (CLSC). This method employs twin agents to establish a closed-loop framework for policy and sampling data distribution, gradually improving policy efficiency while ensuring policy safety. Base on theoretical and experimental analysis, CLSC elucidates the performance, influencing factors, and advantages of this approach in policy safety scenarios. These findings provide novel research insights and practical directions for advancing safe RL. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Node transfer with graph contrastive learning for class-imbalanced node classification

引用

Neural Networks 2025年 190卷 107674页

作者： Li, Yangding Zhao, Xiangchao Zeng, Yangyang Feng, Hao Chai, Jiawei Xie, Hao Fu, Shaobin Zhang, Shichao College of Information Science and Engineering Hunan Normal University Changsha China Hunan Provincial Laboratory of Intelligent Computing and Language Information Processing Hunan Normal University Changsha China Guangxi Key Lab of Multi-Source Information Mining and Security Guangxi Normal University Guilin China

In graph representation learning, the class imbalance problem is a significant challenge that has received much attention from academics. Although current approaches have shown promising results, they have not adequately addressed the problems of node quantity imbalance and feature space imbalance in datasets. This research presents a node transfer with graph contrastive learning framework (NT-GCL) that aims to improve the representation capabilities of graph neural networks for minority classes nodes by balancing node quantity and feature space distributions. First, the proposed node transfer algorithm redistributes misclassified nodes from majority classes to achieve a balanced distribution of node quantity and feature space. This approach effectively prevents the feature space of minority classes from being compressed by majority classes during information propagation, further mitigating potential imbalance issues. Subsequently, the self-supervised contrastive learning strategy is employed to train the model without relying on labels, reducing the bias introduced by labeled data. Experiments conducted with various encoders on six public datasets demonstrate that NT-GCL exhibits strong competitiveness in class-imbalanced node classification. © 2025 Elsevier Ltd

关键词： Class imbalance Graph contrastive learning Graph neural network Node classification Node transfer

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：