Диссертация (Методы, алгоритмы и программные средства распознавания русской телефонной спонтанной речи), страница 23
Описание файла
Файл "Диссертация" внутри архива находится в папке "Методы, алгоритмы и программные средства распознавания русской телефонной спонтанной речи". PDF-файл из архива "Методы, алгоритмы и программные средства распознавания русской телефонной спонтанной речи", который расположен в категории "". Всё это находится в предмете "технические науки" из Аспирантура и докторантура, которые можно найти в файловом архиве СПбГУ. Не смотря на прямую связь этого архива с СПбГУ, его также можно найти и в других разделах. , а ещё этот архив представляет собой кандидатскую диссертацию, поэтому ещё представлен в разделе всех диссертаций на соискание учёной степени кандидата технических наук.
Просмотр PDF-файла онлайн
Текст 23 страницы из PDF
— P. 257–286.32. Baum, L. Statistical inference for probabilistic functions of finite state Markovchains [Text] / L. Baum, T. Petrie // Ann. Math. Statist. — 1966. — Vol. 37,no. 6. — P. 1554––1563.33. Dempster, A. Maximum-likelihood from incomplete data via the EM algorithm[Text] / A. Dempster, N.
Laird, D. Rubin // J. Roy. Stat. Soc. Ser. B. — 1977. —Vol. 39, no. 1. — P. 1––38.34. Gopinath, R. Maximum likelihood modeling with Gaussian distributions for classification [Text] / R. Gopinath // Proc. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). — 1998. — Vol. 2. — P. 661––664.35. Gales, M. Semi-tied Covariance Matrices for Hidden Markov Models [Text] /M. Gales // IEEE Trans.
in Speech and Audio Processing. — 1999. — Vol. 7. —P. 272––281.36. Leggetter, C. Maximum likelihood linear regression for speaker adaptation of continuous density hidden Markov models [Text] / C. Leggetter, P. Woodland // Comput. Speech Lang. — 1995. — Vol. 9, no. 2. — P. 171––185.37. Gales, M. Maximum likelihood linear transformations for HMM-based speechrecognition [Text] / M. Gales // Comput. Speech Lang. — 1998. — Vol. 12,no. 2.
— P. 75––98.38. Chesta, C. Maximum a posteriori linear regression for hidden markov model adaptation [Text] / C. Chesta, O. Siohan, C.-H. Lee // proc. Eurospeech. — 1999. —P. 211––214.13739. Povey, D. Discriminative Training for Large Vocabulary Speech Recognition[Text] : Ph. D. thesis / D. Povey ; Cambridge University Engineering Dept. —Cambridge : [s. n.], 2003.40. Deep neural networks for acoustic modeling in speech recognition: the sharedviews of four research groups [Text] / G. Hinton, L.
Deng, D. Yu [et al.] // IEEESignal Process. Mag. — 2012. — Vol. 29, no. 6. — P. 82––97.41. Rumelhart, D. Learning representations by back-propagating errors [Text] /D. Rumelhart, G. Hinton, R. Williams // Nature. — 1986. — Vol. 323. — P. 533––536.42. Bengio, Y. Practical recommendations for gradient-based training of deep architectures [Text] / Y.
Bengio // Neural Networks: Tricks of the Trade. — 2012. —P. 437––478.43. Efficient backprop [Text] / Y. LeCun, L. Bottou, G. Orr, K. Müller // Neural Networks: Tricks of the Trade. — 1998. — P. 9––50.44. Bottou, L. Online Algorithms and Stochastic Approximations [Text] / L. Bottou //Online Learning and Neural Networks / Ed. by D. Saad. — Cambridge, UK : Cambridge University Press, 1998. — revised, oct 2012.
URL: http://leon.bottou.org/papers/bottou-98x (online; accessed: 02.02.2016).45. Nesterov, Y. A method of solving a convex programming problem with convergence rate O(1/k 2 ) [Text] / Y. Nesterov // Sov. Math. Dokl. — 1983. —Vol. 27. — P. 372––376.46. Sutskever, I. Training Recurrent Neural Networks [Text] : Ph. D. thesis /I. Sutskever ; CS Dept., University of Toronto.
— Toronto : [s. n.], 2012.47. Bengio, Y. Advances in optimizing recurrent networks [Text] / Y. Bengio,N Boulanger-Lewandowski, R. Pascanu // Proc. IEEE International Conference onAcoustics, Speech and Signal Processing (ICASSP). — 2013. — P. 8624–8628.48. Snoek, J. Practical Bayesian optimization of machine learning [Text] / J.
Snoek,H. Larochelle, R. Adams // arXiv preprint arXiv:1206.2944. — 2012. — URL:http://arxiv.org/abs/1206.2944 (online; accessed: 02.02.2016).13849. Veselý, K. Convolutive Bottleneck Network Features for LVCSR [Text] /K. Veselý, M. Karafiat, F. Grezl // Proc. IEEE Workshop on Automatic SpeechRecognition and Understanding (ASRU). — 2011. — P. 42–47.50. Nielsen, M. Neural Networks and Deep Learning [Electronic resource]. —2016. — URL: http://neuralnetworksanddeeplearning.com/ (online; accessed:02.02.2016).51.
Hinton, G. A fast learning algorithm for deep belief nets [Text] / G. Hinton,S. Osindero, Y. Teh // Neural Comput. — 2006. — Vol. 18. — P. 1527–1554.52. Smolensky, P. Information Processing in Dynamical Systems: Foundations ofHarmony Theory [Text] / P. Smolensky. — Boulder : Department of ComputerScience, University of Colorado, 1986. — P. 194–281.53. Hinton, G. Training products of experts byminimizing contrastive divergence[Text] / G. Hinton // Neural Comput. — 2002.
— Vol. 14, no. 8. — P. 1771–1800.54. Hinton, G. A practical guide to training restricted Boltzmann machines. TechnicalReport UTML TR 2010-003 [Text] / G. Hinton. — [S. l.] : University of Toronto,2010.55. Greedy layer-wise training of deep networks [Text] / Y. Bengio, P. Lamblin,D. Popovici, H. Larochelle // Proc. Neural Information Processing Systems (NIPS)Workshop. — 2006. — P. 153––160.56. Extracting and composing robust features with denoising autoencoders [Text] /P.
Vincent, H. Larochelle, Y. Bengio, P. Manzagol // Proc. International Conference on Machine Learning (ICML). — 2008. — P. 1096––1103.57. Trentin, E. A survey of hybrid ANN/HMMmodels for automatic speech recognition [Text] / E. Trentin, M. Gori // Neurocomputing. — 2001. — Vol. 37, no. 1. —P. 91––126.58. Bourlard, H. Links between Markov models and multilayer perceptrons [Text] /H. Bourlard, C. Wellekens // IEEE Trans. Pattern Anal. Mach. Intell.
(PAMI). —1990. — Vol. 12, no. 12. — P. 1167––1178.13959. Bourlard, H. Continuous speech recognition using multilayer perceptrons withhidden Markov models [Text] / H. Bourlard, C. Wellekens // Proc. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). —1990. — P. 413––416.60. Morgan, N. Neural networks for statistical recognition of continuous speech[Text] / N. Morgan, H. Bourlard // Proc. IEEE.
— 1995. — Vol. 83, no. 5. —P. 742––772.61. Large vocabulary continuous speech recognition with context-dependent DBNHMMs [Text] / G. Dahl, D. Yu, L. Deng, A. Acero // Proc. IEEE InternationalConference on Acoustics, Speech and Signal Processing (ICASSP). — 2011. —P. 4688––4691.62. Context-dependent pre-trained deep neural networks for large-vocabulary speechrecognition [Text] / G. Dahl, D. Yu, L. Deng, A. Acero // IEEE Trans.
Audio,Speech Lang. Process. — 2012. — Vol. 20, no. 1. — P. 30––42.63. Kingsbury, B. Scalable minimum bayes risk training of deep neural networkacousticmodels using distributed hessian-free optimization [Text] / B. Kingsbury,T. Sainath, H. Soltau // Proc. Annual Conference of International Speech Communication Association (INTERSPEECH). — 2012.64. Seide, F.
Conversational speech transcription using context-dependent deep neural networks [Text] / F. Seide, G. Li, D. Yu // Proc. Annual Conference of International Speech Communication Association (INTERSPEECH). — 2011. —P. 437––440.65. Yu, D. Roles of pre-training and fine-tuning in context-dependent DBN-HMMsfor real-world speech recognition [Text] / D. Yu, L. Deng, G. Dahl // Proc.
NeuralInformation Processing Systems (NIPS) Workshop. — 2010.66. GMM-free DNN training [Text] / A. Senior, G. Heigold, M. Bacchiani, H. Liao //Proc. IEEE International Conference on Acoustics, Speech and Signal Processing(ICASSP). — 2014.
— P. 5602–5606.67. Maximum mutual information estimation of hidden markov model parametersfor speech recognition [Text] / L. Bahl, P. Brown, P. De Souza, R. Mercer //140Proc. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). — 1986. — Vol. 11. — P. 49–52.68. Kapadia, S. MMI training for continuous phoneme recognition on the TIMITdatabase [Text] / S. Kapadia, V. Valtchev, S. Young // Proc.
IEEE InternationalConference on Acoustics, Speech and Signal Processing (ICASSP). — 1993. —Vol. 2. — P. 491–494.69. Boosted MMI for model and feature-space discriminative training [Text] /D. Povey, D. Kanevsky, B. Kingsbury [et al.] // Proc. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). — 2008.
— P. 4057–4060.70. Povey, D. Minimum phone error and I-smoothing for improved discriminativetraining [Text] / D. Povey, P. Woodland // Proc. IEEE International Conferenceon Acoustics, Speech and Signal Processing (ICASSP). — 2008. — Vol. 1. —P. 105–108.71. Goel, V. Minimum Bayes-risk automatic speech recognition [Text] / V. Goel,W. Byrne // Comput. Speech Lang. — 2000.