检索结果-内蒙古大学图书馆

COMPUTER-ASSISTED ONLINE LEARNING OF ENGLISH ORAL PRONUNCIATION BASED ON DAE END-TO-END RECURRENT NEURAL NETWORKS

SCALABLE COMPUTING-PRACTICE AND EXPERIENCE 2024年第6期25卷 4493-4506页

作者： Lai, Kangsheng Mo, Liujun Pingxiang Univ Foreign Languages Coll Pingxiang 337000 Jiangxi Peoples R China

With the development of globalization, learning a second language has received increasing attention from people. To improve English oral proficiency, a computer-aided online learning system for English oral pronunciation is studied. A denoising autoenco der is integrated into the system to create a simplified end-to-end recurrent neural network for pronunciation detection and diagnosis based on deep learning. The study first collected and preprocessed oral pronunciation data of English learners, including enhancing speech signals and reducing noise. Next, an RNN model with Long Short-Term Memory (LSTM) as the core was constructed to capture time series characteristics in pronunciation. And use DAE to extract features and reduce the influence of background noise to enhance the recognition of pronunciation features. At the same time, the study utilized web crawler technology to collect a large amount of oral pronunciation data from non-native English learners, and constructed an English oral corpus containing pronunciation errors. And in order to simulate real situations, white noise and pink noise were artificially added to the corpus in the study, and they were divided into training and testing sets in a ratio of 60% to 40%. The results showed that the classification accuracy of the system in the training and testing sets under white noise environment was 78.97% and 94.01%, respectively, and the classification accuracy in the pink noise environment was 76.19% and 94.03%, respectively. The system's error detection accuracy in vowel and consonant pronunciation detection is 88.91% and 91.68%, respectively, and the error correction accuracy in vowel and consonant pronunciation detection is 90.67% and 91.96%, respectively. In summary, the research on computer-aided online learning of English oral pronunciation based on denoising Auto Encoders end-to-end recurrent neural networks has effectively improved learning efficiency.

关键词： denoising autoencoder Spoken English pronunciation Recurrent neural network End-to-end

来源：评论

学校读者我要写书评

暂无评论

Comparative analysis of the denoising effect of unstructured vs. convolutional autoencoders

Comparative analysis of the denoising effect of unstructured...

引用

Emerging Trends in Smart Technologies (ICETST), 2020 International Conference on

作者： Sajid Majeed Yusra Mansoor Sana Qabil Farooq Majeed Behraj Khan National University of Computer and Emerging Sciences Karachi Pakistan

ISBN: (数字)9781728171135

ISBN: (纸本)9781728171142

Adding noise to data affects the prediction of a discriminator. Some like a deep neural network extract features directly from inputs, where the quality of the features may be affected by the amount of the noise. A deep neural model specifically meant for feature extraction is the autoencoder, while it has also been extended to perform denoising. In this paper, we investigate the denoising effect of an encoder on different nature as well as different amounts of additive noise. The experiments are evaluated on a linearized autoencoder as well as a convolutional autoencoder, which is especially meant for image data. denoising autoencoders (DAE) and Convolutional denoising autoencoders (CDAE) are evaluated by introducing with Gaussian, Salt and Pepper, and Poisson types of noise with a factor of 0.5. The results shows 0.12, 0.09, 0.47 Mean Squared Error (MSE) for DAE and 0.13, 0.10 and 0.9 MSE in case of CDAE with the same amount of noise factor added, alluding to the insight that a lack of focus on structure in the model may help it focus more on the denoising task.

关键词： Image denoising CNN denoising autoencoder convolutional denoising autoencoder Noise reduction image denoising Mean square error Noise figure Noise ENCODER

来源：评论

学校读者我要写书评

暂无评论

Robust Automatic Recognition of Speech with background music

Robust Automatic Recognition of Speech with background music

引用

IEEE International Conference on Acoustics, Speech and Signal Processing

作者： Jiri Malek Jindrich Zdansky Petr Cerva Faculty of Mechatronics Informatics and Interdisciplinary Studies Technical University of Liberec Studentska 2 461 17 Czech Republic

ISBN: (纸本)9781509041183

This paper addresses the task of Automatic Speech Recognition (ASR) with music in the background, where the accuracy of recognition may deteriorate significantly. To improve the robustness of ASR in this task, e.g. for broadcast news transcription or subtitles creation, we adopt two approaches: 1) multi-condition training of the acoustic models and 2) denoising autoencoders followed by acoustic model training on the preprocessed data. In the latter case, two types of autoencoders are considered: the fully connected and the convolutional network. Presented experimental results show that all the investigated techniques are able to improve the recognition of speech distorted by music significantly. For example, in the case of artificial mixtures of speech and electronic music (low Signal-to-Noise Ratio (SNR) of 0 dB), we achieved absolute improvement of accuracy by 35.8%. For real-world broadcast news and a high SNR (about 10 dB), we achieved improvement by 2.4%. The important advantage of the studied approaches is that they do not deteriorate the accuracy in scenarios with clean speech (the decrease is about 1%).

关键词： Robust recognition background music feature enhancement denoising autoencoder multi-condition training Speech Music Speech recognition Acoustic models Recognition (Psychology) aquifer storage and recovery high SNR signal to noise ratios Electronic music ELECTRONIC INSTRUMENTS

来源：评论

学校读者我要写书评

暂无评论

Open Set Recognition by Regularising Classifier with Fake Data Generated by Generative Adversarial Networks

Open Set Recognition by Regularising Classifier with Fake Da...

引用

IEEE International Conference on Acoustics, Speech and Signal Processing

作者： Inhyuk Jo Jungtaek Kim Hyohyeong Kang Yong-Deok Kim Seungjin Choi Department of Computer Science and Engineering POSTECH Korea Software R&D Center Device Solutions Samsung Electronics Korea

ISBN: (纸本)9781538646595

We present a new method to generate fake data in unknown classes in generative adversarial networks (GANs) framework. The generator in GANs is trained to generate somewhat similar to data in known classes but the different one by modelling noisy distribution on feature space of a classifier using proposed marginal denoising autoencoder. The generated data are treated as fake instances in unknown classes and given to the classifier to make it be robust to the real unknown classes. Our results show that synthetic data can act as fake unknown classes and keep down the certainty of the classifier on real unknown classes meanwhile the classification capability of known classes is not degenerated, even improved.

关键词： Generative adversarial networks denoising autoencoder Open set recognition Feature matching Classifiers Unknown Open set Eigenvector synthetic data Classes Network

来源：评论

学校读者我要写书评

暂无评论

ROBUST RECOGNITION OF SPEECH WITH BACKGROUND MUSIC IN ACOUSTICALLY UNDER-RESOURCED SCENARIOS

ROBUST RECOGNITION OF SPEECH WITH BACKGROUND MUSIC IN ACOUST...

引用

IEEE International Conference on Acoustics, Speech and Signal Processing

作者： Jiri Malek Jindrich Zdansky Petr Cerva Faculty of Mechatronics Informatics and Interdisciplinary Studies Technical University of Liberec Studentská 2 461 17 Liberec Czech Republic

ISBN: (纸本)9781538646595

This paper addresses the task of Automatic Speech Recognition (ASR) with music in the background. We consider two different situations: 1) scenarios with very small amount of labeled training utterances (duration 1 hour) and 2) scenarios with large amount of labeled training utterances (duration 132 hours). In these situations, we aim to achieve robust recognition. To this end we investigate the following techniques: a) multi-condition training of the acoustic model, b) denoising autoencoders for feature enhancement and c) joint training of both above mentioned techniques. We demonstrate that the considered methods can be successfully trained with the small amount of labeled acoustic data. We present substantially improved performance compared to acoustic models trained on clean speech. Further, we show a significant increase of accuracy in the under-resourced scenario, when utilizing additional amount of non-labeled data. Here, the non-labeled dataset is used to improve the accuracy of the feature enhancement via autoencoders. Subsequently, the autoencoders are jointly fine-tuned along with the acoustic model using the small amount of labeled utterances.

关键词： robust speech recognition feature enhancement denoising autoencoder multi-condition training joint training Acoustic models Speech recognition Speech Multiple signal classification Functional training

来源：评论

学校读者我要写书评

暂无评论

Deep Difference Representation Learning for Multi-spectral Imagery Change Detection

Deep Difference Representation Learning for Multi-spectral I...

引用

2016 5th International Conference on Advanced Materials and Computer Science(ICAMCS 2016)

作者： Hui Zhang Puzhao Zhang Department of Integrated Circuit Design and Integrated System School of MicroelectronicsXidian University Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education of China Xidian University

Change detection is an ongoing hot topic in multi-spectral imagery applications,how to exploit the available spectral information effectively for change detection is still an open *** the noise interference and redundancy of multi-spectral imagery,it is important and necessary to learn more abstract and robust feature from raw spectrums for change detection *** this paper,a deep difference representation learning model is proposed for multi-spectral change *** this model,two stacked denoising autoencoders are established,one for learning more abstract features from raw spectrums blocks,and the other for learning difference representations from the stacked change *** former is used to weaken noise interference and reduce redundancy,while the latter has the ability to highlight changes and suppress unchanged *** experimental results on real multi-spectral data demonstrate the feasibility,effectiveness and robustness of the proposed deep difference representation learning model on multi-spectral change detection task.

关键词： Change Detection Multi-Spectral Imagery Difference Representation denoising autoencoder Deep Learning

来源：评论

学校读者我要写书评

暂无评论

Online Anomaly Detection with Streaming Data based on Fine-grained Feature Forecasting

Online Anomaly Detection with Streaming Data based on Fine-g...

引用

第33届中国控制与决策会议

作者： Keying Liu Wentao Mao Huadong Shi Chao Wu Jiaxian Chen School of Computer and Information Engineering Henan Normal University

In the industrial applications like fault diagnosis and health management,monitoring data generally reaches sequentially in a streaming *** recognize fault occurrence in real time without system halt,it is necessary to improve the accuracy and stability of anomaly detection with streaming *** solve this problem,a new online anomaly detection method with streaming data is proposed based on fine-grained feature ***,to conduct fine-grained decomposition of features,a denoising autoencoder network is run to extract multiple-dimensional deep features of online data in the initial period of normal ***,a forecasting model with tensor Tucker decomposition and ARIMA is conducted to predict the fluctuation trend of all feature ***,the deviation degree between the prediction values and sequentially-arrived data is calculated,and an alarm threshold is built according to the 95%confidence interval of the maximum *** the anomalous state data can be detected in real *** paper adopts the problem of bearing early fault online detection as an example,and run comparative experiments on the IEEE PHM Challenge 2012 bearing *** results show that the proposed method has good detection accuracy and is with no false alarm,while the model training does not rely on any offline *** the proposed method is applicable to the problem of online anomaly detection

关键词： Streaming data denoising autoencoder Anomaly detection Alarm threshold ARIMA model

来源：评论

学校读者我要写书评

暂无评论

Vibration Based Multimodal Feature Learning Framework For Wind Turbine Gearbox Fault Diagnosis

Vibration Based Multimodal Feature Learning Framework For Wi...

引用

The 29th International Congress on Condition Monitoring and Diagnostic Engineering Management

作者： Guoqian Jiang Ping Xie Xiao Wang Qun He Meng Chen School of Electrical Engineering Yanshan University

Fault diagnosis of wind turbine gearboxes has attracted great attention,and it is still a quite challenging task to extract effective features from the raw vibration signals to achieve accurate and reliable fault ***,various vibration based features from time,frequency and time-frequency representation are extracted for further fault recognition achieved by *** different representations can be viewed as a modality of the vibration *** different modalities can provide complementary information for diagnosis,and may have different sensitivities to health ***,there is need to fuse them to enhance fault diagnosis *** by the idea of multimodal deep learning(MDL),this paper presents a multimodal vibration feature learning and fusion framework,with the aim to fuse multimodal homologous features derived from the single channel gearbox vibration signal using deep autoencoder *** framework mainly includes unsupervised compact vibration feature learning from each modality and unified representation learning by fusing learned multimodal features followed by a simple classifier for fault *** on a wind turbine gearbox test rig show that the proposed framework achieves best diagnosis accuracy compared to single modality based approaches.

关键词： wind turbine gearbox vibration signal fault diagnosis multimodal deep learning deep autoencoders denoising autoencoder

来源：评论

学校读者我要写书评

暂无评论

MULTI-TASK autoencoder FOR NOISE-ROBUST SPEECH RECOGNITION

MULTI-TASK AUTOENCODER FOR NOISE-ROBUST SPEECH RECOGNITION

引用

IEEE International Conference on Acoustics, Speech and Signal Processing

作者： Haoyi Zhang Conggui Liu Nakamasa Inoue Koichi Shinoda Tokyo Institute of Technology Japan

ISBN: (纸本)9781538646595

For speech recognition in noisy environments, we propose a multi-task autoencoder which estimates not only clean speech features but also noise features from noisy speech. We introduce the deSpeeching autoencoder, which excludes speech signals from noisy speech, and combine it with the conventional denoising autoencoder to form a unified multi-task autoencoder (MTAE). We evaluate it using the Aurora 2 dataset and CHIME 3 dataset. It reduced WER by 15.7% from the conventional denoising autoencoder in the Aurora 2 test set A.

关键词： denoising autoencoder deSpeeching autoencoder robust speech recognition Speech recognition Noise Speech auroras Noise reduction Chimes Dataset

来源：评论

学校读者我要写书评

暂无评论

SEPARATION MATRIX OPTIMIZATION USING ASSOCIATIVE MEMORY MODEL FOR BLIND SOURCE SEPARATION

SEPARATION MATRIX OPTIMIZATION USING ASSOCIATIVE MEMORY MODE...

引用

European Signal Processing Conference

作者： Motoi Omachi Tetsuji Ogawa Tetsunori Kobayashi Masaru Fujieda Kazuhiro Katagiri Department of the computer science Waseda University Oki Electric Industry Co. Ltd.

ISBN: (纸本)9781479988518

A source signal is estimated using an associative memory model (AMM) and used for separation matrix optimization in linear blind source separation (BSS) to yield high quality and less distorted speech. Linear-filtering-based BSS, such as independent vector analysis (IVA), has been shown to be effective in sound source separation while avoiding non-linear signal distortion. This technique, however, requires several assumptions of sound sources being independent and generated from non-Gaussian distribution. We propose a method for estimating a linear separation matrix without any assumptions about the sources by repeating the following two steps: estimating non-distorted reference signals by using an AMM and optimizing the separation matrix to minimize an error between the estimated signal and reference signal. Experimental comparisons carried out in simultaneous speech separation suggest that the proposed method can reduce the residual distortion caused by IVA.

关键词： Convolutional neural network denoising autoencoder Associative memory model Linear filtering Blind source separation

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：