检索结果-内蒙古大学图书馆

21st IEEE International Conference on Machine Learning and Applications (IEEE ICMLA)

作者： Chhetri, Amit S. Hilmes, Philip Athi, Mrudula Shankar, Nikhil Amazon Inc Seattle WA 98109 USA

ISBN: (纸本)9781665462839

In this paper, we present the design of a robust deep neural network based speech enhancement (DNNSE) solution for joint noise reduction and dereverberation under real-world acoustic conditions. This makes our proposed solution suitable for smart-speaker products that encounter a wide variety of acoustic challenges during their real-world deployment. We provide a systematic introduction to the acoustic challenges involved in real-world products and perform a detailed analysis to compare DNNSE models using metrics such as short-time objective intelligibility (STOI), scale-invariant signal-to-distortion-ratio (SI-SDR), and generalizability to unseen acoustic conditions. We then develop a robust DNNSE solution along with a robust training procedure that are well suited for the acoustic challenges specified in the paper. Through detailed analysis, we demonstrate that our DNNSE solution performs and generalizes better than a baseline solution that is 5x-6x times larger.

关键词： Deep neural network speech enhancement densely connected network convolutional recurrent network convolutional encoder-decoder

来源：评论

学校读者我要写书评

暂无评论

Physics-constrained deep learning for data assimilation of subsurface transport

引用

Energy and AI 2021年第1期3卷 104-114页

作者： Haiyi Wu Rui Qiao Department of Mechanical Engineering Virginia TechBlacksburgVA 24061USA

Data assimilation of subsurface transport is important in many energy and environmental applications,but its solution is typically *** this work,we build physics-constrained deep learning models to predict the full-scale hydraulic conductivity,hydraulic head,and concentration fields in porous media from sparse measure-ment of these *** model is developed based on convolutional neural networks with the encoding-decoding *** model is trained by minimizing a loss function that incorporates residuals of governing equations of subsurface transport instead of using labeled *** trained,the model predicts the unknown conductivity,hydraulic head,and concentration fields with an average relative error<10%when the data of these observables is available at 12.2%of the grid points in the porous *** model has a robust predictive performance for porous media with different conductivities and transport under different Péclet number(0.5

关键词： Physics-constrained deep learning Data assimilation Subsurface transport convolutional encoder-decoder

来源：评论

学校读者我要写书评

暂无评论

Control framework for collaborative robot using imitation learning-based teleoperation from human digital twin to robot digital twin*

引用

MECHATRONICS 2022年第0期85卷

作者： Lee, Hyunsoo Kim, Seong Dae Amin, Mohammad Aman Ullah Al Kumoh Natl Inst Technol Sch Ind Engn Gumi South Korea Univ Tennessee Chattanooga Dept Engn Management & Technol Chattanooga TN USA Univ Texas Arlington Dept Ind Engn Arlington TX USA 615 McCallie Ave Chattanooga TN 37403 USA

Despite the deployment of collaborative robots for various industrial processes, their teaching and control remain comparatively difficult tasks compared with general industrial robots. Various imitation learning methods involving the transfer of human poses to a collaborative robot have been proposed. However, most of these methods depend heavily on deep learning-based human recognition algorithms that fail to recognize complicated human poses. To address this issue, we propose an automated/semi-automated vision-based teleoperation framework using human digital twin and a collaborative robot digital twin models. First, a human pose is recognized and reasoned to a human skeleton model using a convolution encoder-decoder architecture. Next, the developed human digital twin model is taught using the skeletons. As human and collaborative robots have different joints and rotation architectures, pose mapping is achieved using the proposed Bezier curve-based smooth approximation. Then, a real collaborative robot is controlled using the developed robot digital twin. Furthermore, the proposed framework works successfully using a human digital twin in the case of recognition failures of human poses. To verify the effectiveness of the proposed framework, transfers of several human poses to a real collaborative robot are tested and analyzed.

关键词： Collaborative robot Teleoperation framework Imitation learning Digital twin Bezier curve-based smooth pose mapping convolutional encoder-decoder

来源：评论

学校读者我要写书评

暂无评论

Gated dynamic convolutions with deep layer fusion for abstractive document summarization

引用

COMPUTER SPEECH AND LANGUAGE 2021年 66卷 101159-101159页

作者： Kwon, Hongseok Go, Byung-Hyun Park, Juhong Lee, Wonkee Jeong, Yewon Lee, Jong-Hyeok Pohang Univ Sci & Technol Dept Comp Sci & Engn 77 Cheongam Ro Pohang 37673 South Korea Pohang Univ Sci & Technol Grad Sch Artificial Intelligence Pohang South Korea

We present a novel abstractive document summarization based on the recently proposed dynamic convolutional encoder-decoder architectures. We address several aspects of summarization that are not well modeled by the basic architecture, by integrating multiple layers of the encoder, controlling information flow in the hierarchy, and exploiting external knowledge. First, we propose a simple and efficient deep layer fusion to extract salient information from the encoder layers. Second, we propose a gating mechanism to control and maintain important contextual information through the encoder-decoder layers into dynamic convolutions. Lastly, we put part-of-speech information into the model as external knowledge to better predict filters for dynamic convolutions. We evaluate our model using ROUGE metrics on three different datasets: CNN-DM, NEWSROOM-ABS, and XSUM. Experimental results show that the proposed model outperforms the state-of-the-art abstractive models on NEWSROOM-ABS and XSUM and shows comparable scores on CNN-DM. (C) 2020 Elsevier Ltd. All rights reserved.

关键词： Document summarization Gated dynamic convolutions Deep layer fusion convolutional encoder-decoder Text generation

来源：评论

学校读者我要写书评

暂无评论

Optimized and autonomous machine learning framework for characterizing pores, particles, grains and grain boundaries in microstructural images

引用

COMPUTATIONAL MATERIALS SCIENCE 2021年 196卷 110524-110524页

作者： Perera, Roberto Guzzetti, Davide Agrawal, Vinamra Auburn Univ Dept Aerosp Engn Auburn AL 36849 USA

Additively manufactured metals exhibit heterogeneous microstructure which dictates their material and failure properties. Experimental microstructural characterization techniques generate a large amount of data that requires expensive computational resources. In this work, an optimized machine learning (ML) framework is proposed to autonomously and efficiently characterize pores, particles, grains and grain boundaries (GBs) from a given microstructure image. First, using a classifier convolutional Neural Network (CNN), defects such as pores, powder particles, or GBs were recognized from a given microstructure. Depending on the type of defect, two different processes were used. For powder particles or pores, binary segmentations were generated using an optimized convolutional encoder-decoder Network (CEDN). The binary segmentations were used to obtain particle and pore size and location using an object detection ML network (YOLOv5). For GBs, another optimized CEDN was developed to generate RGB segmentation images, which were used to obtain grain size distribution using two regression CNNs. To optimize the RGB CEDN, the Deep Emulator Network SEarch (DENSE) method which employs the Covariance Matrix Adaptation - Evolution Strategy (CMA-ES) was implemented. The optimized RGB segmentation network showed a substantial reduction in training time and GPU usage compared to the unoptimized network, while maintaining high accuracy. Lastly, the proposed framework showed a significant improvement in analysis time when compared to conventional methods.

关键词： Machine learning Microstructural feature characterization convolutional encoder-decoder convolutional neural networks Deep Emulator Network SEarch (DENSE)

来源：评论

学校读者我要写书评

暂无评论

REDUNDANT convolutional NETWORK WITH ATTENTION MECHANISM FOR MONAURAL SPEECH ENHANCEMENT

REDUNDANT CONVOLUTIONAL NETWORK WITH ATTENTION MECHANISM FOR...

引用

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

作者： Lan, Tian Lyu, Yilan Hui, Guoqiang Mokhosi, Refuoe Li, Sen Liu, Qiao Univ Elect Sci & Technol China Sch Informat & Software Engn Chengdu Sichuan Peoples R China CETC Big Data Res Inst Co Ltd Guiyang Peoples R China

ISBN: (纸本)9781509066315

The redundant convolutional encoder-decoder network has proven useful in speech enhancement tasks. It can capture localized time-frequency details of speech signals through both the fully convolutional network structure and feature selection capability resulting from the encoder-decoder mechanism. However, it does not explicitly consider the signal filtering mechanism, which we regard as important for speech enhancement models. In this study, we introduce an attention mechanism into the convolutional encoder-decoder model. This mechanism adaptively filters channel-wise feature responses by explicitly modeling attentions (on speech versus noise signals) between channels. Experimental results show that the proposed attention model is effective in capturing speech signals from background noise, and performs especially better in unseen noise conditions compared to other state-of-the-art models.

关键词： Squeeze-and-Excitation convolutional encoder-decoder speech enhancement fully convolutional network attention mechanism

来源：评论

学校读者我要写书评

暂无评论

Modeling the dynamics of PDE systems with physics-constrained deep auto-regressive networks

引用

JOURNAL OF COMPUTATIONAL PHYSICS 2020年 403卷 1页

作者： Geneva, Nicholas Zabaras, Nicholas Univ Notre Dame Ctr Informat & Computat Sci 311 Cushing Hall Notre Dame IN 46556 USA

In recent years, deep learning has proven to be a viable methodology for surrogate modeling and uncertainty quantification for a vast number of physical systems. However, in their traditional form, such models can require a large amount of training data. This is of particular importance for various engineering and scientific applications where data may be extremely expensive to obtain. To overcome this shortcoming, physics-constrained deep learning provides a promising methodology as it only utilizes the governing equations. In this work, we propose a novel auto-regressive dense encoder-decoder convolutional neural network to solve and model non-linear dynamical systems without training data at a computational cost that is potentially magnitudes lower than standard numerical solvers. This model includes a Bayesian framework that allows for uncertainty quantification of the predicted quantities of interest at each time-step. We rigorously test this model on several non-linear transient partial differential equation systems including the turbulence of the Kuramoto-Sivashinsky equation, multi-shock formation and interaction with 1D Burgers' equation and 2D wave dynamics with coupled Burgers' equations. For each system, the predictive results and uncertainty are presented and discussed together with comparisons to the results obtained from traditional numerical analysis methods. (C) 2019 Elsevier Inc. All rights reserved.

关键词： Physics-informed machine learning Auto-regressive model Deep neural networks convolutional encoder-decoder Uncertainty quantification Dynamic partial differential equations

来源：评论

学校读者我要写书评

暂无评论

SEPARATED NOISE SUPPRESSION AND SPEECH RESTORATION: LSTM-BASED SPEECH ENHANCEMENT IN TWO STAGES

SEPARATED NOISE SUPPRESSION AND SPEECH RESTORATION: LSTM-BAS...

引用

IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA)

作者： Strake, Maximilian Defraene, Bruno Fluyt, Kristoff Tirry, Wouter Fingscheidt, Tim Tech Univ Carolo Wilhelmina Braunschweig Inst Commun Technol D-38106 Braunschweig Germany NXP Semicond Prod Line Voice & Audio Solut B-3001 Leuven Belgium

ISBN: (纸本)9781728111230

Regression based on neural networks (NNs) has led to considerable advances in speech enhancement under non-stationary noise conditions. Nonetheless, speech distortions can be introduced when employing NNs trained to provide strong noise suppression. We propose to address this problem by first suppressing noise and subsequently restoring speech with specifically chosen NN topologies for each of these distinct tasks. A mask-estimating long short-term memory (LSTM) network is employed for noise suppression, while the speech restoration is performed by a fully convolutional encoder-decoder (CED) network, where we introduce temporal modeling capabilities by using a convolutional LSTM layer in the bottleneck. We show considerable performance gains over reference methods of up to 0.26 MOS points (PESQ) and the ability to significantly improve intelligibility in terms of STOI for low-SNR conditions.

关键词： speech enhancement two-stage processing long short-term memory convolutional encoder-decoder

来源：评论

学校读者我要写书评

暂无评论

Front-end speech enhancement for commercial speaker verification systems

引用

SPEECH COMMUNICATION 2018年 99卷 101-113页

作者： Eskimez, Sefik Emre Soufleris, Peter Duan, Zhiyao Heinzelman, Wendi Univ Rochester 500 Wilson Blvd Rochester NY 14627 USA Voice Biometr Grp 12 Penns Trail Newtown PA 18966 USA

Commercial speaker verification systems are an important component in security services for various domains, such as law enforcement, government, and finance. These systems are sensitive to noise present in the input signal, which leads to inaccurate verification results and hence security breaches. Traditional speech enhancement (SE) methods have been employed to improve the performance of speaker verification systems. However, to the best of our knowledge, the impact of state-of-the-art speech enhancement techniques has not been analyzed for text-independent automatic speaker verification (ASV) systems using real-world utterances. In this work, our contribution is twofold. First, we propose two deep neural network (DNN) architectures for SE, and we compare the performance of the proposed networks with the existing work. We evaluate the resulting SE networks using the objective measures of perceptual evaluation of speech quality (PESQ) and short-time objective intelligibility (STOI). Second, we analyze the performance of ASV systems when SE methods are used as frontend processing to remove the non-stationary background noise. We compare the resulting equal error rate (EER) using our DNN based SE approaches, as well as existing SE approaches, with real customer data and the freely available RedDots dataset. Our results show that our DNN based SE approaches provide benefits for speaker verification performance.

关键词： Speech enhancement Automatic speaker verification Deep neural networks Bidirectional LSTM convolutional encoder-decoder

来源：评论

学校读者我要写书评

暂无评论

Improving Spatial Context in CNNs for Semantic Medical Image Segmentation 4

Improving Spatial Context in CNNs for Semantic Medical Image...

引用

4th IAPR Asian Conference on Pattern Recognition (ACPR)

作者： Mesbah, Russel McCane, Brendan Mills, Steven Robins, Anthony Univ Otago Dept Comp Sci Dunedin New Zealand

ISBN: (纸本)9781538633540

convolutional Neural Networks (CNNs) have been widely used in the semantic segmentation of medical images. Current CNN-based approaches don't fully exploit information about the local neighbourhood of the pixels being classified thus the inferred labels can be affected by the scope of the information included in the sampling frame. Due to the scarcity of the labelled medical images, using the entire image as the encoder-decoder's input frame is practically unattainable. Thus, random sampling using small frames is an indispensable part of the network training. Furthermore, the average pixel-wise accuracy gives the average likelihood of correct classification across all the pixels in the frame where this likelihood is actually not the same for every pixel. We propose a new approach to address these issues by using multiple neighbourhoods around the pixel of interest and aggregating different hypotheses about the pixel's label. The results produced by this method are comparable with the state of the art solutions. In addition, the method is capable of detecting less accurate regions by assessing the consistency of labelling while shifting the sampling frame across these pixels.

关键词： Semantic Medical Image Segmentation CNN convolutional encoder-decoder

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：