Semi-Orthogonal Low-Rank Matrix Factorization for Deep Neural Networks

Keywords

Matrix decomposition, Factorization, Computer science, Artificial neural network, Rank (graph theory)

References

• Training Very Deep NetworksRupesh K. Srivastava, Klaus Greff, Jürgen Schmidhuber — 2015
• The Kaldi Speech Recognition ToolkitDaniel Povey — 2011
• Understanding the difficulty of training deep feedforward neural networksXavier Glorot, Yoshua Bengio — 2010
• Mean-normalized stochastic gradient for large-scale deep learningSimon Wiesler, Alexander Richard, Ralf Schlüter — 2014
• Deep Residual Learning for Image RecognitionKaiming He, Xiangyu Zhang, Shaoqing Ren — 2015
• Restructuring of deep neural network acoustic models with singular value decompositionJian Xue, Jinyu Li, Yifan Gong — 2013
• On the compression of recurrent neural networks with an application to LVCSR acoustic modeling for embedded speech recognitionRohit Prabhavalkar, Ouais Alsharif, Antoine Bruguier — 2016
• Pronunciation and silence probability modeling for ASRGuoguo Chen, Hainan Xu, Minhua Wu — 2015
• A time delay neural network architecture for efficient modeling of long temporal contextsVijayaditya Peddinti, Daniel Povey, Sanjeev Khudanpur — 2015
• Audio augmentation for speech recognitionTom Ko, Vijayaditya Peddinti, Daniel Povey — 2015
• Model Compression Applied to Small-Footprint Keyword SpottingGeorge Tucker, Minhua Wu, Ming Sun — 2016
• Purely Sequence-Trained Neural Networks for ASR Based on Lattice-Free MMIDaniel Povey, Vijayaditya Peddinti, Daniel Gálvez — 2016
• Low Latency Acoustic Modeling Using Temporal Convolution and LSTMsVijayaditya Peddinti, Yiming Wang, Daniel Povey — 2018
• An Exploration of Dropout with LSTMsGaofeng Cheng, Vijayaditya Peddinti, Daniel Povey — 2017
• Deep Learning-Based Telephony Speech Recognition in the WildKyu J. Han, Seongjun Hahm, Byung‐Hak Kim — 2017
• A Pruned Rnnlm Lattice-Rescoring Algorithm for Automatic Speech RecognitionHainan Xu, Tongfei Chen, Dongji Gao — 2018
• Neural Network Language Modeling with Letter-Based Features and Importance SamplingHainan Xu, Ke Li, Yiming Wang — 2018
• Highway long short-term memory RNNS for distant speech recognitionYu Zhang, Guoguo Chen, Dong Yu — 2015

Cited by

• Evaluating Code-Switched Malay-English Speech Using Time Delay Neural Networks.Soma Khan, Madhab Pal, Joyanta Basu — 2018
• Training Neural Speech Recognition Systems with Synthetic Speech AugmentationJason Li, Ravi Teja Gadde, Boris Ginsburg — 2018
• The Airbus Air Traffic Control speech recognition 2018 challenge: towards ATC automatic transcription and call sign detectionThomas Pellegrini, Jérôme Farinas, Estelle Delpech — 2018
• The Marchex 2018 English Conversational Telephone Speech Recognition SystemSeongjun Hahm, Iroro Orife, Shane Walker — 2018
• Multiple beamformers with ROVER for the CHiME-5 ChallengeSining Sun, Yangyang Shi, Ching-Feng Yeh — 2018
• The NWPU System for CHiME-5 ChallengeZhao Zhiwei, Jian Wu, Lei Xie — 2018
• The SHNU system for the CHiME-5 ChallengeYanhua Long, Renke He — 2018
• JHU Diarization System DescriptionZili Huang, Leibny Paola Garcia, Jesús Villalba — 2018
• 探討鑑別式訓練聲學模型之類神經網路架構及優化方法的改進 (Discriminative Training of Acoustic Models Leveraging Improved Neural Network Architecture and Optimization Method) [In Chinese]Wei-Cheng Chao, Hsiu-Jui Chang, Tien-Hong Lo — 2018
• Improving LF-MMI Using Unconstrained Supervisions for ASRHossein Hadian, Daniel Povey, Hossein Sameti — 2018
• The VOiCES from a Distance Challenge 2019 Evaluation PlanMahesh Kumar Nandwana, Julien van Hout, Mitchell McLaren — 2019
• Optimisation methods for training deep neural networks in speech recognitionAdnan Haider — 2019
• Acoustic Modeling for Overlapping Speech Recognition: Jhu Chime-5 Challenge SystemVimal Manohar, Szu-Jui Chen, Zhiqi Wang — 2019
• Bidirectional LSTM with Extended Input ContextGaofeng Cheng, Lu Huang, Jiasong Sun — 2018
• Lattice-based lightly-supervised acoustic model trainingJoachim Fainberg, Ondřej Klejch, Steve Renals — 2019
• Semi-supervised acoustic model training for five-lingual code-switched ASRAstik Biswas, Emre Yılmaz, Febe de Wet — 2019
• Improved low-resource Somali speech recognition by semi-supervised acoustic and language model trainingAstik Biswas, Raghav Menon, Ewald van der Westhuizen — 2019
• Lattice-Based Unsupervised Test-Time Adaptation of Neural Network Acoustic ModelsOndřej Klejch, Joachim Fainberg, Peter Bell — 2019
• ShrinkML: End-to-End ASR Model Compression Using Reinforcement LearningŁukasz Dudziak, Mohamed S. Abdelfattah, Ravichander Vipperla — 2019
• Acoustic Modeling for Automatic Lyrics-to-Audio AlignmentChitralekha Gupta, Emre Yılmaz, Haizhou Li — 2019

Related papers

• Which scale for parton densities? Highlight from k_t-factorizationB. Guiot — 2018
• Some rank equalities of some outer inverses of the same matrixTao Xie — 2009
• Spectral Factorization of Rational Matrix Valued FunctionsGiacomo Baggio — 2014
• Generalization of real interval matrices to other fieldsElena Rubei — 2018
• On computing normalized coprime factorization of arbitrary rational matricesCristian Oarâ, Vlad Ionescu — 2000
• Existence of a low rank or ℋ︁‐matrix approximant to the solution of a Sylvester equationLars Grasedyck — 2004
• Construction of a full row-rank matrix system for multiple scanning directions in discrete tomographyXiezhang Li, James Diffenderfer, Jiehua Zhu — 2017