Filters








237 Hits in 5.5 sec

Native Language Identification in Very Short Utterances using Bidirectional Long Short-Term Memory Network

Farah Adeeba, Sarmad Hussain
2019 IEEE Access  
The bidirectional long short-term memory (BLSTM) neural networks are adopted for the classification of utterances among the native languages.  ...  INDEX TERMS Native language identification, BLSTM, RNN, Urdu L2.  ...  BIDIRECTIONAL LSTM RNN MODEL Long short term memory (LSTM) network [34] is a special type of recurrent neural network (RNN) with the capability of learning long-term dependencies.  ... 
doi:10.1109/access.2019.2896453 fatcat:ar7c5nw7ifewhbmrkauha3ldz4

Part of speech tagging: a systematic review of deep learning and machine learning approaches

Alebachew Chiche, Betselot Yitagesu
2022 Journal of Big Data  
Furthermore, the presence of ambiguity when tagging terms with different contextual meanings inside a sentence cannot be overlooked.  ...  Then, recent trends and advancements of DL and ML-based part-of-speech-taggers are presented in terms of the proposed approaches deployed and their performance evaluation metrics.  ...  Long short-term memory A Long Short-Term Memory (LSTM) is a special kind of RNN network architecture, which has the capability of learning long-term dependencies.  ... 
doi:10.1186/s40537-022-00561-y fatcat:fpmxdnm76benxms6wckwiqykpy

Protein-Protein Interaction Article Classification Using a Convolutional Recurrent Neural Network with Pre-trained Word Embeddings

Sérgio Matos, Rui Antunes
2017 Journal of Integrative Bioinformatics  
In this work we used a convolutional recurrent neural network for identifying relevant articles for extracting information regarding protein interactions.  ...  pharmacological action for example.  ...  Acknowledgement This work was supported by Portuguese National Funds through FCT -Foundation for Science and Technology, through a FCT Investigator grant (IF/01694/2013) and through Project IF/01694/2013  ... 
doi:10.1515/jib-2017-0055 pmid:29236678 fatcat:t4jobrn63jafddar6tgyrsrsrm

Layer Trajectory LSTM

Jinyu Li, Changliang Liu, Yifan Gong
2018 Interspeech 2018  
This layer-LSTM scans the outputs from time-LSTMs, and uses the summarized layer trajectory information for final senone classification.  ...  However, an LSTM-RNN with too many vanilla LSTM layers is very hard to train and there still exists the gradient vanishing issue if the network goes too deep.  ...  Introduction Recently, significant progress has been made in automatic speech recognition (ASR) when switching from the deep neural networks (DNNs) [1] to recurrent neural networks (RNNs) with long short-term  ... 
doi:10.21437/interspeech.2018-1485 dblp:conf/interspeech/LiLG18 fatcat:2gqbzkeegrafji6xsbfmewmy74

Layer Trajectory LSTM [article]

Jinyu Li, Changliang Liu, Yifan Gong
2018 arXiv   pre-print
This layer-LSTM scans the outputs from time-LSTMs, and uses the summarized layer trajectory information for final senone classification.  ...  However, an LSTM-RNN with too many vanilla LSTM layers is very hard to train and there still exists the gradient vanishing issue if the network goes too deep.  ...  Introduction Recently, significant progress has been made in automatic speech recognition (ASR) when switching from the deep neural networks (DNNs) [1] to recurrent neural networks (RNNs) with long short-term  ... 
arXiv:1808.09522v1 fatcat:qbphfpcfyjhsnlwbokkvi24jhe

A Review on Methods and Applications in Multimodal Deep Learning [article]

Jabeen Summaira, Xi Li, Amin Muhammad Shoib, Jabbar Abdul
2022 arXiv   pre-print
Deep Learning has implemented a wide range of applications and has become increasingly popular in recent years.  ...  Despite the extensive development made for unimodal learning, it still cannot cover all the aspects of human learning.  ...  This framework transformed the text into speech from voices using a short-shifting memory buffer. J. Shen et al. [88] proposed "Tacotron2.  ... 
arXiv:2202.09195v1 fatcat:wwxrmrwmerfabbenleylwmmj7y

Speech Enhancement with Phase Sensitive Mask Estimation using a Novel Hybrid Neural Network

Mojtaba Hasannezhad, Zhiheng Ouyang, Wei-Ping Zhu, Benoit Champagne
2021 IEEE Open Journal of Signal Processing  
Index Terms-speech enhancement, convolutional neural network, grouped long short-term memory, attention technique, phase sensitive mask.  ...  In this paper, we propose a hybrid neural network model integrating a new low-complexity fully-convolutional CNN and a long shortterm memory (LSTM) network, a variation of RNN, to estimate a phase-sensitive  ...  Since a traditional RNN acts like an FC network with an infinite number of hidden layers and thus suffers from the vanishing and exploding gradient problem, long short-term memory (LSTM) networks were  ... 
doi:10.1109/ojsp.2021.3067147 fatcat:27orae2fhnebblzcvcs4vamyzq

Particle Filter Recurrent Neural Networks

Xiao Ma, Peter Karkus, David Hsu, Wee Sun Lee
2020 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
: while an RNN relies on a long, deterministic latent state vector, a PF-RNN maintains a latent state distribution, approximated as a set of particles.  ...  For effective learning, we provide a fully differentiable particle filter algorithm that updates the PF-RNN latent state distribution according to the Bayes rule.  ...  Specifically, we propose PF-LSTM and PF-GRU, the particle filter extensions of Long Short Term Memory (LSTM) (Hochreiter and Schmidhuber 1997) and Gated Recurrent Unit (GRU) (Cho et al. 2014) .  ... 
doi:10.1609/aaai.v34i04.5952 fatcat:ahea2lhbyfc3phrw2gaesy4evi

StableEmit: Selection Probability Discount for Reducing Emission Latency of Streaming Monotonic Attention ASR [article]

Hirofumi Inaguma, Tatsuya Kawahara
2021 arXiv   pre-print
While attention-based encoder-decoder (AED) models have been successfully extended to the online variants for streaming automatic speech recognition (ASR), such as monotonic chunkwise attention (MoChA)  ...  As a result, the scale of the selection probabilities is increased, and the values can reach a threshold for token emission earlier, leading to a reduction of emission latency and deletion errors.  ...  Experimental evaluations with Long Short-Term Memory (LSTM) and causal Conformer [14] encoders show that the proposed method significantly reduces the deletion errors and the emission latency of MoChA  ... 
arXiv:2107.00635v2 fatcat:y76yindi7jfzbcgzfmrqgjjgya

Exploring Layer Trajectory LSTM with Depth Processing Units and Attention

Jinyu Li, Liang Lu, Changliang Liu, Yifan Gong
2018 2018 IEEE Spoken Language Technology Workshop (SLT)  
time-LSTM layer, and uses the summarized layer trajectory information for final senone classification.  ...  In this paper, we extend our recently proposed layer trajectory LSTM (ltL-STM) and present a generalized framework, which is equipped with a depth processing block that scans the hidden states of each  ...  (RNNs) with long short-term memory (LSTM) units [6] .  ... 
doi:10.1109/slt.2018.8639637 dblp:conf/slt/LiLLG18 fatcat:os7y3obsxreyhggv5f6euuowuu

A Hierarchical Approach for Generating Descriptive Image Paragraphs

Jonathan Krause, Justin Johnson, Ranjay Krishna, Li Fei-Fei
2017 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
for an image.  ...  In this paper we overcome these limitations by generating entire paragraphs for describing images, which can tell detailed, unified stories.  ...  Alternative recurrent architectures such as long-short term memory (LSTM) [9] help alleviate this problem through a gating mechanism that improves gradient flow.  ... 
doi:10.1109/cvpr.2017.356 dblp:conf/cvpr/KrauseJKF17 fatcat:dbye5dz2t5huho7mijnjm7tjsa

A Hierarchical Approach for Generating Descriptive Image Paragraphs [article]

Jonathan Krause, Justin Johnson, Ranjay Krishna, Li Fei-Fei
2017 arXiv   pre-print
for an image.  ...  In this paper we overcome these limitations by generating entire paragraphs for describing images, which can tell detailed, unified stories.  ...  Alternative recurrent architectures such as long-short term memory (LSTM) [9] help alleviate this problem through a gating mechanism that improves gradient flow.  ... 
arXiv:1611.06607v2 fatcat:yvmtd3usbvbwdjclends7lzqri

Applicable Predictive Maintenance Diagnosis Methods in Service-Life Prediction of District Heating Pipes

Pakdad Pourbozorgi Langroudi, Ingo Weidlich
2020 Environmental and Climate Technologies  
In recent years, District heating (DH) in the countries that employing this technology broadly, turned to a vital energy infrastructure for delivering heat from suppliers to the consumers.  ...  The transition from reactive maintenance to proactive maintenance have improved a lot the reliability to the system.  ...  ACKNOWLEDGEMENT The authors acknowledge the financial support by the Federal Ministry for Economic Affairs and Energy of Germany in the project Instandhaltung-FW (project number 03ET1625B).  ... 
doi:10.2478/rtuect-2020-0104 fatcat:leykgvlbi5chrpwhmrochkkdfq

Forecasting of the Prevalence of Dementia Using the LSTM Neural Network in Taiwan

Stephanie Yang, Hsueh-Chih Chen, Chih-Hsien Wu, Meng-Ni Wu, Cheng-Hong Yang
2021 Mathematics  
The present study established a model based on the architecture of the long short-term memory (LSTM) neural network for predicting the number of dementia cases in Taiwan, which considers the effects of  ...  The LSTM network is a variant of recurrent neural networks (RNNs), which possesses a special gate structure and avoids the problems in RNNs of gradient explosion, gradient vanishing, and long-term memory  ...  Acknowledgments: We would like to thank the reviewers for their valuable comments, which help us to improve our paper a lot. Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/math9050488 fatcat:tfrxwpuzhbcm5nn2fu5z2k7eqq

A Systematic Review on Affective Computing: Emotion Models, Databases, and Recent Advances [article]

Yan Wang, Wei Song, Wei Tao, Antonio Liotta, Dawei Yang, Xinlei Li, Shuyong Gao, Yixuan Sun, Weifeng Ge, Wei Zhang, Wenqiang Zhang
2022 arXiv   pre-print
Next, we survey and taxonomize state-of-the-art unimodal affect recognition and multimodal affective analysis in terms of their detailed architectures and performances.  ...  Affective computing plays a key role in human-computer interactions, entertainment, teaching, safe driving, and multimedia integration.  ...  Deep RNN learning for TSA. RNN-based TSA is capable of processing long sequence data.  ... 
arXiv:2203.06935v3 fatcat:h4t3omkzjvcejn2kpvxns7n2qe
« Previous Showing results 1 — 15 out of 237 results