Edmonton / Dallas / Makuhari / Landsdowne / Kyoto / Philadelphia / Stroudsburg / Atlanta / Portland / Vancouver / Prague / Washington / Lyon / /
Company
Toshiba Research Europe Ltd / /
Country
Barbados / France / Japan / Canada / Czech Republic / / /
Facility
College Park / /
IndustryTerm
speech recognition systems / generalized lattice rescoring algorithm / confusion network / neural networks / recurrent network / recurrent neural network / conversational web data / 1k-best and 10k-best rescoring systems / generated network / beam search / above history clustering algorithm / above baseline systems / lattice processing tools / conversational telephone speech transcription task / downstream applications / neural network / baseline systems / back propagation algorithm / beam search decoders / conversational telephone speech / web text sources / speech processing / large vocabulary conversational telephone speech recognition task / pseudo-code algorithm / expansion algorithm / search space / /
MusicAlbum
Speech / /
Organization
Cambridge University / US Government / U.S. Securities and Exchange Commission / Cambridge Research Lab. / A. Deoras / T. Mikolov / S. Kombrink / and K. Church / US Federal Reserve / /
Person
G. E. Hintont / D. E. Rumelhart / R. J. Williams / J. J. Odell / V / / /
Position
RT / times RT / LDC Fisher / Fisher / /
ProvinceOrState
Texas / Virginia / Hawaii / Pennsylvania / Georgia / /
EFFICIENT LATTICE RESCORING USING RECURRENT NEURAL NETWORK LANGUAGE MODELS X. Liu, Y. Wang, X. Chen, M. J. F. Gales & P. C. Woodland Cambridge University Engineering Dept, Trumpington St., Cambridge, CB2 1PZ U.K. Email: