Microsoft Research Asia / Microsoft Corporation / Xerox College Publishing / ADAPTATION OF CONTEXT-DEPENDENT DEEP NEURAL NETWORKS / Blaisdell Publishing Company / /
Country
United States / China / /
Currency
USD / /
Facility
Prentice Hall / Carnegie Mellon University / Tsinghua University / Johns Hopkins University / /
IndustryTerm
speaker-independent neural network / Online Service Division / deep belief networks / contrast divergence algorithm / neural network / linear hidden network / fine-tuning process using back-propagation algorithm / adaptation algorithms / linear input network / back-propagation algorithm / shallow networks / voice search data / neural networks / distant talking voice search / filter-bank cepstral coefficients / context-dependent deep neural networks / context-dependent deep-neural-network / hybrid systems / linear output network / /
MusicAlbum
SPEECH / /
Organization
Carnegie Mellon University / Johns Hopkins University / Tsinghua University / Beijing / U.S. Securities and Exchange Commission / /
Person
Yu-Chi Ho / B. Kingsbury / P. Nguyen / Arthur Earl Bryson / V. Vanhoucke / T. Sainath / /
Position
speaker / /
Product
Microsoft Corporation Portable Audio Device / /
RadioStation
DNN / /
Technology
SPEECH RECOGNITION / contrast divergence algorithm / neural network / adaptation algorithms / back-propagation algorithm / fine-tuning process using back-propagation algorithm / speech recognition system / /
ADAPTATION OF CONTEXT-DEPENDENT DEEP NEURAL NETWORKS FOR AUTOMATIC SPEECH RECOGNITION Kaisheng Yao1 , Dong Yu2 , Frank Seide3 , Hang Su3,4 , Li Deng2 , and Yifan Gong1 1 Online Service Division, Microsoft Corporation, R