D12-1031 work has grown around the task of machine transliteration . In this task , the goal is
D13-1021 training set to train statistical machine transliteration model for our base - line . The
D09-1111 goal is to measure the impact of machine transliterations on end-to-end translation quality
D09-1069 " Can Chinese phonemes improve machine transliteration ? " Actually , this is the second
D13-1021 Japanese-to-English statistical machine transliteration in patent domain using patent
D09-1111 Finally , we have demonstrated that machine transliteration is immediately useful to end-to-end
D13-1021 corpora . It enables the statistical machine transliteration to be bootstrapped from bilingual
C04-1089 proposed a probabilistic model for machine transliteration . In this model , a word in the
C04-1103 reduce the development efforts of machine transliteration systems and improve accuracy
C04-1103 propose a unified framework for machine transliteration , direct orthographical mapping
D09-1111 features . Finally , we show that machine transliterations have a positive impact on machine
C04-1103 present a novel framework for machine transliteration . The new framework carries out
D12-1003 Karimi et al. , 2011 ) . Although machine transliteration works particularly well for proper
D13-1021 use in bootstrapping statistical machine transliteration using Japanese-to-English patent
C04-1089 translation . We use a variant of the machine transliteration method proposed by ( Knight and
C04-1103 . 2 Previous Work The topic of machine transliteration has been studied extensively
D13-1021 and named entities . Statistical machine transliteration ( Knight and Graehl , 1998 )
D13-1021 for bootstrapping statistical machine transliteration from automatically extracted
C04-1103 orthographical mapping ( DOM ) for machine transliteration and back-transliteration . Under
D11-1089 their primary goal is to build a machine transliteration system or to build a bilingual
隱藏細節