检索结果-内蒙古大学图书馆

End-to-End Streaming Customizable keyword Spotting Based on Text-Adaptive Neural Search 1

18th National Conference on Man-Machine Speech Communication, NCMMSC 2023

作者： Yang, Baochen Guo, Jiaqi Li, Haoyu Xi, Yu Zhuo, Qing Yu, Kai X-LANCE Lab Department of Computer Science and Engineering MoE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University Shanghai China State Key Laboratory of Media Convergence Production Technology and Systems Beijing China AISpeech Ltd. Suzhou China Department of Automation Tsinghua University Beijing China

ISBN: (数字)9789819706013

ISBN: (纸本)9789819706006

Streaming keyword spotting (KWS) is an important technique for voice assistant wake-up. While KWS with a preset fixed keyword has been well studied, test-time customizable keyword spotting in streaming mode remains a great challenge due to the lack of pre-collected keyword-specific training data and the requirement of streaming detection output. In this paper, we propose a novel end-to-end text-adaptive neural search architecture with a multi-label trigger mechanism to allow any pre-trained ASR acoustic model to be effectively used for fast streaming customizable keyword spotting. Evaluation results on various datasets show that our approach significantly outperforms both traditional post-processing baseline and the neural search baseline, meanwhile achieving a 44x search speedup compared to the traditional post-processing method. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Customizable keyword keyword spotting Streaming Text-adaptive search

来源：评论

学校读者我要写书评

暂无评论

Constructing multi-level urban clusters based on population distributions and interactions

arXiv

引用

arXiv 2022年

作者： Cao, Wenpu Dong, Lei Cheng, Ying Wu, Lun Guo, Qinghua Liu, Yu Institute of Remote Sensing and Geographic Information Systems School of Earth and Space Sciences Peking University Beijing100871 China State Key Laboratory of Media Convergence Production Technology and Systems Beijing China

A city (or an urban cluster) is not an isolated spatial unit, but a combination of areas with closely linked socio-economic activities. However, so far, we lack a consistent and quantitative approach to define multi-level urban clusters through these socio-economic connections. Here, using granular population distribution and flow data from China, we propose a bottom-up aggregation approach to quantify urban clusters at multiple spatial scales. We reveal six ‘phases’ (i.e., levels) in the population density-flow diagram, each of which corresponds to a spatial configuration of urban clusters from large to small. Besides, our results show that Zipf’s law appears only after the fifth level, confirming the spatially dependent nature of urban laws. Our approach does not need pre-defined administrative boundaries and can be applied effectively on a global scale. © 2022, CC BY.

关键词： Big data

来源：评论

学校读者我要写书评

暂无评论

Single-Shot Cascade Bounding Box Refinement Neural Network for High Quality Object Detection

Single-Shot Cascade Bounding Box Refinement Neural Network f...

引用

Chinese Control and Decision Conference, CCDC

作者： Qiong Wu Yi Fang Fei Long Qiang Ling Anhui JiangHuai Automobile Group CO. Ltd. Hefei China University of Science and Technology of China Hefei China Chinaso Inc. Beijing China State Key Laboratory of Media Convergence Production Technology and Systems Xinhua News Agency Beijing China

This paper proposes a novel one-stage detection method with cascade bounding box refinement, called CasDet. Generally one-stage detection methods generate less accurate bounding boxes than two-stage methods. Motivated by the location refinement of bounding boxes of two-stage methods, we present a cascade bounding box refinement for one-stage detection networks. By refining the detection results for multiple times, CasDet can generate high quality bounding boxes. To further improve the feature representation, we present a multi-scale feature aggregation structure (MBFA). By fusing features with adjacent three scales and adding multiple top-down and bottom-up aggregation paths, MBFA introduces rich contextual information and local details into each detection feature. The experimental results on COCO benchmark confirm that our CasDet outperforms existing state-of-the-art detection methods and achieves a better balance between speed and accuracy.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Comparative Study of Cross-Sentence Features for Named Entity Recognition

A Comparative Study of Cross-Sentence Features for Named Ent...

引用

Innovations and Development of Information Technologies and Robotics (IDITR), International Conference on

作者： Sheng-Fu Wang Jing Huang Baohua Zhang Jia Li School of Computer Science and Technology Beijing Institute of Technology Beijing China Huajian eTone Language Technology (Beijing) Co. Ltd State Key Laboratory of Media Convergence Production Technology and Systems China Xinhua New Media Culture Communication Co. Ltd Big Data&Artificial Intelligence Laboratory Industrial and Commercial Bank of China Shanghai China

Recently, a growing number of Named Entity Recognition (NER) methods utilize cross-sentence features (also known as contexts) to improve the performance of NER models, instead of using single-sentence information alone. As far as we know, most NER models choose to exploit pre- and post-sentences to capture cross-sentence features. Generally, current NER studies focus only on the model architecture to capture better token representations. However, there is no in-depth exploration on how to better model cross-sentence features. In this paper, based on the span classification model, we investigate the effect of cross-sentence features under different settings. Specifically, we evaluate the impact of context stitching, context window size, context window padding, and classifier token of pre-trained language model (PLM) on model performance. Comparative experimental results show that appropriate incorporation of document-level contexts can considerably improve the NER metrics. Furthermore, we find that several factors can be used to improve the performance of NER models: (1) use domain-specific PLMs, but not classifier tokens; (2) use only preceding contexts for generic text, and random contexts for specialized text; (3) truncate overly long contexts when the context window is small, and preserve sentence integrity when the window is large; (4) set the context window size to about 200 for the basic size PLM.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Zoned Mapping Network from SDR Video to HDR Video

Zoned Mapping Network from SDR Video to HDR Video

引用

IEEE Advanced Information Management,Communicates,Electronic and Automation Control Conference (IMCEC)

作者： Rui Tang Fang Meng Lefu Bai Communication University of China Beijing China The Technical Bureau of Xinhua News Agency Beijing China Xinhua News Agency State Key Laboratory of Media Convergence Production Technology and Systems Beijing China

ISBN: (纸本)9781665479691

HDR video is popular for its brilliant colors, incredible brightness and rich details. Considering the strict production conditions of HDR video, most videos are currently saved in SDR format. Converting massive SDR videos into HDR format through technical methods can quickly fill the vacancy of HDR videos in the market and improve the utilization efficiency of film and television resources. In this paper, we propose a zoned mapping network to convert the video format. According to SDR OETF, the video frames are firstly segmented into three regions, highlight, medium and darkness. Then a tone mapping model is designed and respectively trained for the above three regions. Moreover, a detail enhancement model is proposed to form a composite inverse tone mapping network. Experiments show that our method can not only achieve good subjective visual quality, but also accomplish excellent results on various objective metrics, such as PSNR, SSIM, VIF, etc.

关键词： Measurement Visualization TV Automation Brightness production Color

来源：评论

学校读者我要写书评

暂无评论

Spatially-varying Warping for Panoramic Image Stitching

Spatially-varying Warping for Panoramic Image Stitching

引用

第34届中国控制与决策会议

作者： Jiabing Xiong Feng Li Fei Long Yonghua Xu Song Wang Jun Xu Qiang Ling Department of Automation University of Science and Technology of China State Key Laboratory of Media Convergence Production Technology and Systems Xinhua News Agency Chinaso Inc. Hefei Softec Auto Electronic Co.LTD

Panoramic image stitching is widely implemented to many application scenarios,such as satellite remote sensing detection,street view mapping and medical image *** this paper,we study two processes that have great impact on the visual effects of the stitched images and are the most concerned by researchers,namely image registration and stitching seam *** contrast to the traditional panoramic image stitching algorithms that use only one global projection transformation matrix for image alignment,this paper proposes a spatially-varying deformation method that uses multiple local transformation matrices,including the APAP(As-Projective-As-Possible) method that can effectively reduce the image alignment error and geometric transformation distortion,the SPHP(Shape-Preserving Half-Projective) method that can combine good alignment accuracy and reduced projection distortion,and the AANAP(Adaptive As-Natural-As-Possible) method that can improve the perspective of non-overlapping regions of the stitched image and good *** image stitching performance can be well improved.

关键词： Panoramic image stitching Image registration Seam removal Spatially-varying warping

来源：评论

学校读者我要写书评

暂无评论

Relation-Aware Multi-hop Reasoning forVisual Dialog 10th

Relation-Aware Multi-hop Reasoning forVisual Dialog

引用

10th CCF Conference on Natural Language Processing and Chinese Computing, NLPCC 2021

作者： Zhao, Yao Chen, Lu Yu, Kai X-LANCE Lab Department of Computer Science and Engineering MoE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University Shanghai China State Key Lab of Media Convergence Production Technology and Systems Beijing China

ISBN: (纸本)9783030884796

Visual dialog is a multi-modal task that requires a dialog agent to answer a series of progressive questions grounded in an image. In this paper, we propose Relation-aware Multi-hop Reasoning Network (i.e. R2N for short) for visual dialog tasks, which can perform multi-hop reasoning during visual co-reference resolution process in a recurrent way. At each hop, in order to fully understand the visual scene in the image, a Relation-aware Graph Attention Network is used, which encodes each image into graphs with multi-type inter-object relations via a graph attention mechanism. Moreover, we find that the auxiliary clustering mechanism on answer candidates is conducive to model’s performance. We evaluate R2N on VisDial v1.0 dataset. Experimental results on the VisDial v1.0 dataset demonstrate that the proposed model is effective and outperforms compared models. © 2021, Springer Nature Switzerland AG.

关键词： Computers

来源：评论

学校读者我要写书评

暂无评论

Enhanced Unsupervised Data Augmentation for Emergency Events Detection and Classification

Enhanced Unsupervised Data Augmentation for Emergency Events...

引用

第33届中国控制与决策会议

作者： Xiaomeng Liu Fei Long Kun Huang Qiang Ling State Key Laboratory of Media Convergence Production Technology and Systems Xinhua News Agency Department of Automation University of Science and Technology of China

We propose to jointly detect and classify emergency events using a multi-class text classifier,which is a typical deep learning architecture with transformer modules and particularly employs Bidirectional Encoder Representations from Transformers(BERT).Deep learning requires a large number of labeled data to ***,deep learning often implements the semi-supervised learning(SSL) method,which is able to use massive unlabeled data to improve performance of supervised deep *** an effective SSL variant,unsupervised data augmentation(UDA) focuses on data augmentation techniques to improve the performance of deep *** present an enhanced version of UDA(EUDA) by mixing more data augmentation strategies and using a problem related *** EUDA targets at emergency event detection and *** that emergency events always have time and location elements,text can be filtered based on this semantic *** propose to use semantic feature aided enhanced unsupervised data augmentation to solve the concerned *** studies on the dataset prepared for the task validates that the proposed EUDA can achieve significantly better performance than supervised learning with a limited size of labeled *** are also carried out on a text classification task,which confirms that EUDA improves performance for BERT neural network.

关键词： Semi-supervised Learning Bidirectional Encoder Representations from Transformers Unsupervised Data Augmentation

来源：评论

学校读者我要写书评

暂无评论

Reinforcement Learning for Extreme Multi-label Text Classification 5th

Reinforcement Learning for Extreme Multi-label Text Classifi...

引用

5th International Conference on Cognitive systems and Signal Processing, ICCSIP 2020

作者： Teng, Hui Li, Yulei Long, Fei Xu, Meixia Ling, Qiang Chinaso Inc. Beijing100077 China State Key Laboratory of Media Convergence Production Technology and Systems Xinhua News Agency Beijing China Department of Computer Science Northeast University Shenyang110819 China Department of Automation University of Science and Technology of China Hefei230027 China

ISBN: (纸本)9789811623356

Extreme multi-label text classification (XMC) is an important yet challenging problem in the NLP community, which refers to the problem of assigning to each document its most relevant subset of class labels from an extremely large label collection. For example, the input text could be a story document on *** and the labels could be story categories that implies the potential meaning. However, naively applying normal neural network models to the XMC problem leads to sub-optimal performance due to the large output space and the label sparsity issue. In this paper, we presents the first attempt at applying reinforcement learning to XMC. Experimental results on public and our own engineering datasets demonstrate that our approach achieves expecting performance compared with the evaluation of the state-of-the-art methods. © 2021, Springer Nature Singapore Pte Ltd.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Event-Triggered Control Strategy Design to Guarantee Desired Control Performance

Event-Triggered Control Strategy Design to Guarantee Desired...

引用

第40届中国控制会议

作者： Shaojie Lan Fei Long keyu Zhan Jie Sheng Feng Li Qiang Ling Dept.of Automation University of Science and Technology of China State Key Laboratory of Media Convergence Production Technology and Systems Xinhua News Agency

This paper considers the problem of exponential practical stabilization of a continuous-time linear time-invariant system in the presence of bounded network delay and process noise. We utilize a model-based periodic event-triggered control strategy which can extract the information contained in the sampling time instants to save quantization bits. Sufficient estimation error conditions to guarantee the prescribed convergence rate are investigated, and appropriate event-triggered control strategies that ensure the desired performance are implemented in this paper. Compared with the time-driven strategies, our method can break the conventional lower bound of the bit rate condition. Simulations are provided to illustrate the proposed strategies.

关键词： exponential stabilization event-triggering networked control systems bit rate condition

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：