检索结果-内蒙古大学图书馆

Physically Consistent Soft-Sensor Development Using sequence-to-sequence Neural Networks

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS 2020年第4期16卷 2829-2838页

作者： Chou, Cheng-Hung Wu, Haibin Kang, Jia-Lin Wong, David Shan-Hill Yao, Yuan Chuang, Yao-Chen Jang, Shi-Shang Ou, John Di-Yi Natl Tsing Hua Univ Dept Chem Engn Hsinchu 30013 Taiwan Natl Taiwan Univ Grad Inst Commun Engn Taipei 10617 Taiwan Natl Yunlin Univ Sci & Technol Dept Chem & Mat Engn Touliu 64002 Yunlin Taiwan Natl Tsing Hua Univ Ctr Energy & Environm Res Hsinchu 30013 Taiwan

Soft sensors attempt to predict the key quality variables that are infrequently available using the sensor and manipulated variables that are readily available. Since only limited amount of labeled data are available, there is always the concern whether the underlying physics were captured so that the model can be reasonably extrapolated. A sequence-to-sequence model in the form of a nonlinear state-observer & x002F;encoder and predictor & x002F;decoder was proposed. The observer can be trained using a large amount of unlabeled data, but in a supervised manner in which the process dynamics is tracked. The encoder output and manipulated variables are used to train the quality predictor. The model is applied to the product impurity predictions of an industrial column. Results show that good predictions and excellent consistency in the sign of estimated gains can be achieved even with limited amount of data. These findings indicated that the proposed sequence-to-sequence data-driven approach is able to capture the underlying physics of the process.

关键词： Data models Temperature measurement Neural networks Feature extraction Predictive models Observers Decoding Gain consistency sequence-to-sequence model soft sensor

来源：评论

学校读者我要写书评

暂无评论

Document Summarization model Based on General Context in RNN

引用

JOURNAL OF INFORMATION PROCESSING SYSTEMS 2019年第6期15卷 1378-1391页

作者： Kim, Heechan Lee, Soowon Soongsil Univ Dept Software Convergence Seoul South Korea Soongsil Univ Sch Software Seoul South Korea

In recent years, automatic document summarization has been widely studied in the field of natural language processing thanks to the remarkable developments made using deep learning models. To decode a word, existing models for abstractive summarization usually represent the context of a document using the weighted hidden states of each input word when they decode it. Because the weights change at each decoding step, these weights reflect only the local context of a document. Therefore, it is difficult to generate a summary that reflects the overall context of a document. To solve this problem, we introduce the notion of a general context and propose a model for summarization based on it. The general context reflects overall context of the document that is independent of each decoding step. Experimental results using the CNN/Daily Mail dataset show that the proposed model outperforms existing models.

关键词： Document Summarization General Context Natural Language Processing sequence-to-sequence model

来源：评论

学校读者我要写书评

暂无评论

Synthesizing waveform sequence-to-sequence to augment training data for sequence-to-sequence speech recognition

引用

ACOUSTICAL SCIENCE AND TECHNOLOGY 2021年第6期42卷 333-343页

作者： Ueno, Sei Mimura, Masato Sakai, Shinsuke Kawahara, Tatsuya Kyoto Univ Grad Sch Informat Sakyo Ku Kyoto 6068501 Japan

sequence-to-sequence (seq2seq) automatic speech recognition (ASR) recently achieves state-of-the-art performance with fast decoding and a simple architecture. On the other hand, it requires a large amount of training data and cannot use text-only data for training. In our previous work, we proposed a method for applying text data to seq2seq ASR training by leveraging text-to-speech (TTS). However, we observe the log Mel-scale filterbank (lmfb) features produced by Tacotron 2-based model are blurry, particularly on the time dimension. This problem is mitigated by introducing the WaveNet vocoder to generate speech of better quality or spectrogram of better time-resolution. This makes it possible to train waveform-input end-to-end ASR. Here we use CNN filters and apply a masking method similar to SpecAugment. We compare the waveform-input model with two kinds of lmfb-input models: (1) lmfb features are directly generated by TTS, and (2) lmfb features are converted from the waveform generated by TTS. Experimental evaluations show the combination of waveform-output TTS and the waveform-input end-to-end ASR model outperforms the lmfb-input models in two domain adaptation settings.

关键词： Speech recognition sequence-to-sequence model Attention-based encoder-decoder model Speech synthesis Data augmentation

来源：评论

学校读者我要写书评

暂无评论

Machine learning based state-of-charge prediction of electrochemical green hydrogen production:Zink-Zwischenschritt-Elektrolyseur(ZZE)

引用

Energy and AI 2024年第2期16卷 269-276页

作者： Daniel Vila Elisabeth Hornberger Christina Toigo STOFF2 GmbH–Gebäude H/Flughafen Tegel 1 13405 BerlinGermany University of Applied Sciences Upper Austria Stelzhamerstraße 23A-4600 WelsAustria Centre for Economic Research on Inclusivity and Sustainability at the University of Galway University RoadGalwayIreland

The intermittency of renewable energy is a key limiting factor for the successful decarbonization of both energy producing and consuming sectors. Green hydrogen has the potential to act as the central energy vector connecting hard-to-abate sectors to renewable power. However, combining energy storage and conversion for a holistic electrolyzer system remains challenging. Here, we show the innovative Zink-Zwischenschritt Elektrolyseur (ZZE), or Zinc Intermediate step Electrolyzer in English, that temporarily decouples the water splitting reaction and uses zinc to store electrical energy in chemical form. To perform optimal operation of a ZZE system, machine learning models were applied to predict the state of charge of a lab scale ZZE system. Using various models, we were able to determine the effectiveness of the prediction and contrast it to state of charge predictions of other energy storage systems. We show that a bi-directional long short-term memory neural network approach has the lowest error within the testing environment. This work serves to perform further ZZE development as well as state of charge prediction for other novel energy storage technologies.

关键词： Electrolyzer Battery ZZE Energy storage State of charge Machine learning sequence-to-sequence model

来源：评论

学校读者我要写书评

暂无评论

A sequence-to-sequence Deep Learning Architecture Based on Bidirectional GRU for Type Recognition and Time Location of Combined Power Quality Disturbance

引用

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS 2019年第8期15卷 4481-4493页

作者： Deng, Yaping Wang, Lu Jia, Hao Tong, Xiangqian Li, Feng Xian Univ Technol Dept Elect Engn Xian 710048 Shaanxi Peoples R China Ningxia Elect Power Res Inst Yinchuan 750011 Peoples R China

In this paper, a sequence-to-sequence deep learning architecture based on the bidirectional gated recurrent unit (Bi-GRU) for type recognition and time location of combined power quality disturbance is proposed. Especially, the proposed methodology can determine the type of each element in input sequence, which is different from existing sequence-to-sequence model employing encoderdecoder network. First, the input sequence is normalized and batched. Second, deep features are extracted frominput sequence by constructing Bi-GRU recurrent neural network, where multiple Bi-GRU layers are stacked together in both forward direction and backward direction. Third, according to aforementioned extracted features, fully connected layer and Softmax are employed to calculate the corresponding probability indicating the category that each element in input sequence is classified to. Fourth, Argmax or Top_K operation is further integrated to determine the type of each element in input sequence by selecting the maximal probability. Finally, the type is recognized, and meanwhile, starting-ending times of disturbances are also located just at the moment when the type is changed. The proposed model is further validated and tested by synthetic signals and practical field signals, respectively. Experimental results demonstrate that the accuracy of type recognition is over 98% for 96 kinds of disturbances including single and combined disturbances with signal-to-noise ration being 20 dB. Besides, the starting-ending times are also located with the absolute error less than six sampling points when sampling frequency is 256 points per cycle with noisy environment.

关键词： Bidirectional gated recurrent unit (Bi-GRU) deep learning power quality disturbance (PQD) sequence-to-sequence model time location type recognition

来源：评论

学校读者我要写书评

暂无评论

Leveraging Causal Inference for Explainable Automatic Program Repair

Leveraging Causal Inference for Explainable Automatic Progra...

引用

IEEE International Conference on Fuzzy Systems (FUZZ-IEEE) / IEEE World Congress on Computational Intelligence (IEEE WCCI) / International Joint Conference on Neural Networks (IJCNN) / IEEE Congress on Evolutionary Computation (IEEE CEC)

作者： Wang, Jianzong Si, Shijing Zhu, Zhitao Qu, Xiaoyang Hong, Zhenhou Xiao, Jing Ping An Technol Shenzhen Co Ltd Shenzhen Peoples R China

ISBN: (数字)9781728186719

ISBN: (纸本)9781728186719

Deep learning models have made significant progress in automatic program repair. However, the black-box nature of these methods has restricted their practical applications. To address this challenge, this paper presents an interpretable approach for program repair based on sequence-to-sequence models with causal inference and our method is called CPR, short for causal program repair. Our CPR can generate explanations in the process of decision making, which consists of groups of causally related input-output tokens. Firstly, our method infers these relations by querying the model with inputs disturbed by data augmentation. Secondly, it generates a graph over tokens from the responses and solves a partitioning problem to select the most relevant components. The experiments on four programming languages (Java, C, Python, and JavaScript) show that CPR can generate causal graphs for reasonable interpretations and boost the performance of bug fixing in automatic program repair.

关键词： Automated Program Repair Program Analysis sequence-to-sequence model Causal Inference Interpretability

来源：评论

学校读者我要写书评

暂无评论

sequence-TO-sequence LABANOTATION GENERATION BASED ON MOTION CAPTURE DATA

SEQUENCE-TO-SEQUENCE LABANOTATION GENERATION BASED ON MOTION...

引用

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

作者： Li, Min Miao, Zhenjiang Ma, Cong Beijing Jiaotong Univ Beijing Peoples R China

ISBN: (纸本)9781509066315

Labanotation is an important notation system for recording dances. Automatically generating Labanotation scores from motion capture data has attracted more interest in recent years. Current methods usually focus on individual movement segments and generate Labanotation symbols one by one. This requires segmenting the captured data sequence in advance. Manual segmentation will consume a lot of time and effort, while automatic segmentation may not be reliable enough. In this paper, we propose a sequence-to-sequence approach that can generate Labanotation scores from unsegmented motion data sequences. First, we extract effective features from motion capture data based on body skeleton analysis. Then, we train a neural network under the encoder-decoder architecture to transform the motion feature sequences to corresponding Labanotation symbols. As such, the dance score is generated. Experiments show that the proposed method performs favorably against state-of-the-art algorithms in the automatic Labanotation generation task.

关键词： Automatic Labanotation generation Motio capture data sequence-to-sequence model

来源：评论

学校读者我要写书评

暂无评论

Attention Forcing for Speech Synthesis 21

Attention Forcing for Speech Synthesis

引用

Interspeech Conference

作者： Dou, Qingyun Efiong, Joshua Gales, Mark J. F. Univ Cambridge Cambridge England

ISBN: (纸本)9781713820697

Auto-regressive sequence-to-sequence models with attention mechanisms have achieved state-of-the-art performance in various tasks including speech synthesis. Training these models can be difficult. The standard approach guides a model with the reference output history during training. However during synthesis the generated output history must be used. This mismatch can impact performance. Several approaches have been proposed to handle this, normally by selectively using the generated output history. To make training stable, these approaches often require a heuristic schedule or an auxiliary classifier. This paper introduces attention forcing, which guides the model with the generated output history and reference attention. This approach reduces the training-evaluation mismatch without the need for a schedule or a classifier. Additionally, for standard training approaches, the frame rate is often reduced to prevent models from copying the output history. As attention forcing does not feed the reference output history to the model, it allows using a higher frame rate, which improves the speech quality. Finally, attention forcing allows the model to generate output sequences aligned with the references, which is important for some down-stream tasks such as training neural vocoders. Experiments show that attention forcing allows doubling the frame rate, and yields significant gain in speech quality.

关键词： sequence-to-sequence model attention mechanism training speech synthesis

来源：评论

学校读者我要写书评

暂无评论

A Conformer-basedWaveform-domain Neural Acoustic Echo Canceller Optimized for ASR Accuracy 23

A Conformer-basedWaveform-domain Neural Acoustic Echo Cancel...

引用

Interspeech Conference

作者： Panchapagesan, Sankaran Narayanan, Arun Shabestary, Turaj Zakizadeh Shao, Shuai Howard, Nathan Park, Alex Walker, James Gruenstein, Alexander Google LLC Mountain View CA 94043 USA

Acoustic Echo Cancellation (AEC) is essential for accurate recognition of queries spoken to a smart speaker that is playing out audio. Previous work has shown that a neural AEC model operating on log-mel spectral features (denoted "logmel" hereafter) can greatly improve Automatic Speech Recognition (ASR) accuracy when optimized with an auxiliary loss utilizing a pre-trained ASR model encoder. In this paper, we develop a conformer-based waveform-domain neural AEC model inspired by the "TasNet" architecture. The model is trained by jointly optimizing Negative Scale-Invariant SNR (SISNR) and ASR losses on a large speech dataset. On a realistic rerecorded test set, we find that cascading a linear adaptive AEC and a waveform-domain neural AEC is very effective, giving 56-59% word error rate (WER) reduction over the linear AEC alone. On this test set, the 1.6M parameter waveform-domain neural AEC also improves over a larger 6.5M parameter logmel-domain neural AEC model by 20-29% in easy to moderate conditions. By operating on smaller frames, the waveform neural model is able to perform better at smaller sizes and is better suited for applications where memory is limited.

关键词： Acoustic echo cancellation TasNet Waveform Neural AEC sequence-to-sequence model multi-task loss

来源：评论

学校读者我要写书评

暂无评论

DATA AUGMENTATION FOR ASR USING TTS VIA A DISCRETE REPRESENTATION

DATA AUGMENTATION FOR ASR USING TTS VIA A DISCRETE REPRESENT...

引用

IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)

作者： Ueno, Sei Mimura, Masato Sakai, Shinsuke Kawahara, Tatsuya Kyoto Univ Grad Sch Informat Sakyo Ku Kyoto Japan

ISBN: (纸本)9781665437394

While end-to-end automatic speech recognition (ASR) has achieved high performance, it requires a huge amount of paired speech and transcription data for training Recently, data augmentation methods have actively been investigated. One method is to use a text-to-speech (TTS) system to generate speech data from text-only data and use the generated speech for data augmentation, but it has been found that the synthesized log Mel-scale filterbank (lmfb) features could have a serious mismatch with the real speech features. In this study, we propose a data augmentation method via a discrete speech representation. The TTS model predicts discrete ID sequences instead of lmfb features, and the ASR also uses the ID sequences as training data. We expect that the use of a discrete representation based on vq-wav2vec not only makes TTS training easier but also mitigates the mismatch with real data. Experimental evaluations show that the proposed method outperforms the data augmentation method using the conventional TTS. We found that it reduces speaker dependency, and the generated features are distributed more closely to the real ones.

关键词： Speech recognition sequence-to-sequence model Data augmentation Vq-wav2vec Speech synthesis

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：