transliteration data 0.003198031
different transliteration 0.00308258
transliteration pairs 0.0028947490000000003
english name 0.0028702800000000002
cantonese transliteration 0.0027865600000000004
machine transliteration 0.0027351840000000003
automatic transliteration 0.0027210460000000004
transliteration datasets 0.002712111
transliteration alternatives 0.0027013590000000004
transliteration units 0.0026986790000000003
acceptable transliteration 0.002675021
forward transliteration 0.002668985
transliteration practice 0.0026666850000000002
transliteration lexicons 0.002656921
chinese characters 0.0024907159999999996
mandarin chinese 0.0024239409999999998
transliteration 0.00239577
chinese sound 0.002389895
english names 0.0023757739999999998
chinese segments 0.002368061
chinese graphemes 0.002337845
chinese homophones 0.0023307669999999996
chinese charac 0.0023299189999999997
chinese fur 0.0023205969999999998
rin chinese 0.0023205969999999998
chinese renditions 0.0023205969999999998
unique english 0.002087905
chinese 0.00205912
english source 0.001976911
source english 0.001976911
certain english 0.001972619
total english 0.001877996
unique name 0.001866745
distinct english 0.001858946
english segments 0.001854661
differences english 0.001840193
name pairs 0.0018235389999999999
english segment 0.001818032
source name 0.001755751
name segmentation 0.0016332
name hilary 0.0016101309999999999
english 0.00154572
tonal language 0.001455067
various alignment 0.001408835
training data 0.001387319
results method 0.001345557
alignment models 0.001314554
test names 0.001279302
alignment preferences 0.0012419269999999999
cantonese names 0.001220844
new names 0.001216405
system performance 0.001208515
mandarin names 0.001194875
cantonese data 0.001193051
personal names 0.001173328
mandarin data 0.0011670819999999998
common set 0.001149021
such judgement 0.00114827
such flexibility 0.00114827
foreign names 0.001135924
method cantonese 0.001112469
language 0.00110621
bigram model 0.001100708
data size 0.001093873
phonetic properties 0.0010847909999999999
average system 0.0010602580000000001
phonetic transcription 0.001053036
different pronunciations 0.00104646
consonant segments 0.001044354
error analysis 0.0010411679999999999
similar candidates 0.001035259
other pho 9.94806E-4
tone information 9.87712E-4
average results 9.86791E-4
soto method 9.864750000000001E-4
finite set 9.78901E-4
alignment 9.77756E-4
depth error 9.68427E-4
good candidates 9.56322E-4
different phono 9.50013E-4
information retrieval 9.48432E-4
unique grapheme 9.39941E-4
grapheme pair 9.34759E-4
preliminary results 9.32229E-4
performance table 9.22005E-4
specific character 9.15651E-4
orthographic mapping 9.02976E-4
grapheme pairs 8.96735E-4
many others 8.95614E-4
grapheme mapping 8.90397E-4
acceptable candidates 8.88902E-4
third character 8.75671E-4
future work 8.5756E-4
pair mappings 8.45059E-4
low frequency 8.366480000000001E-4
cantonese transliterations 8.30536E-4
names 8.30054E-4
important role 8.28795E-4
conventional phoneme 8.282459999999999E-4
related work 8.23807E-4
