View Document Preview and Link Document Date: 2013-10-07 08:11:17 Open Document File Size: 355,51 KB Share Result on Facebook
City Ostendorf / Viterbi / / / Facility University of Edinburgh / / IndustryTerm speech recognition systems / polynomial time algorithms / acoustic model learning algorithm / conversational telephone speech / automatic speech recognition systems / resource management / search space / deep neural networks / recognition systems / / MusicAlbum SPEECH / / Organization University of Edinburgh / Steve Renals Centre for Speech Technology Research / Association for Computational Linguistics / / Person Liang Lu / / Position speaker / Probabilistic pronunciation model / Automatic speech recognition / / ProgrammingLanguage L / J / ML / / PublishedMedium Computational Linguistics / / Technology SPEECH RECOGNITION / Natural Speech Technology / WFST-based EM algorithm / Viterbi-based algorithm / acoustic model learning algorithm / polynomial time algorithms / speech recognition system / EM algorithm / WFSTbased EM algorithm / using the EM algorithm / / SocialTag 
ACOUSTIC DATA-DRIVEN PRONUNCIATION LEXICON FOR LARGE VOCABULARY SPEECH RECOGNITION Liang Lu, Arnab Ghoshal, and Steve Renals Centre for Speech Technology Research, University of Edinburgh, Edinburgh, EH8 9AB, UK {liang.l