Art

J-GLOBAL ID：201402250585086876   Reference number：14A0553060

「Deep Learning(深層学習)」〔第5回〕音声認識のための深層学習

Author (1)： KUBO YOTARO (NTT コミュニケーション科学基礎研)
Material： 人工知能  (人工知能学会誌)
Volume： 29  Issue： 1  Page： 62-71  Publication year： Jan. 01, 2014 
JST Material Number： X0330A  ISSN： 2188-2266  Document type： Article
Article type： 解説  Country of issue： Japan (JPN)  Language： JAPANESE (JA)

Thesaurus term： *speech recognition ,  *Machine Learning ,  *multilayer perceptron ,  neural network (network) ,  hybrid system ,  steepest descent method ,  optimization ,  Gaussian model ,  hidden Markov model ,  *neural network (biomodel) ,  Entropy ,  Information Theory ,  *recurrent neural network ,  *deep learning ,  Cross Entropy ,  modification
Semi thesaurus term： GMM ,  gradient descent method

JST classification (3)： Pattern recognition  (JE07000S) ,  Artificial intelligence  (JE08000Z) ,  Neurocomputers  (JC06010Q)

Reference (32)：

• [Bourlard 94] Bourlard, H. A. and Morgan, N.: Connectionist Speech Recognition: A Hybrid Approach, Vol. 247, Springer (1994)
• [Chen 05] Chen, B. Y., Zhu, Q. and Morgan, N.: Tonotopic multilayered perceptron: A neural network for learning long-term temporal features for speech recognition, Proc. ICASSP-05 (2005)
• [Ghoshal 13] Ghoshal, A., Swietojanski, P. and Renals, S.: Multi- lingual training of deep neural networks, Proc. ICASSP-13 (2013)
• [Graves06] Graves, A., Fernández, S., Gomez, F. and Schmidhuber, J.: Connectionist temporal classification: Labelling unsegmented sequence data with recurrent neural networks, Proc. 23rd Int. Conf. on Machine Learning, pp. 369-376, ACM (2006)
• [Haeb-Umbach 92] Haeb-Umbach, R. and Ney, H.: Linear discriminant analysis for improved large vocabulary continuous speech recognition, ICASSP-92, 1992 IEEE Int. Conf., Vol. 1, pp. 13-16 (1992)

Terms in the title (2)： 深層学習 ,  音声認識