language model 0.0049611099999999995
translation model 0.00429981
model training 0.00350392
reordering model 0.003433428
ing model 0.003228359
acoustic model 0.003207955
segmentation model 0.0032030410000000002
transduction model 0.003164631
model statistics 0.0031606860000000002
tion model 0.0031515980000000002
lation model 0.003076518
tic model 0.003049712
model gcl 0.003048089
guage model 0.003042891
model transformation 0.003039811
model transducer 0.003033899
component model 0.0030297930000000002
model reconstruction 0.0030185950000000002
reconstruction model 0.0030185950000000002
model substan 0.003012891
duction model 0.003012891
language models 0.002922639
training language 0.00289719
model 0.00278392
source language 0.002703042
other language 0.002687878
target language 0.0025812929999999997
language transcription 0.002559736
many language 0.002542545
language pairs 0.002527643
language pair 0.0024978549999999998
language modeling 0.00249654
statistical language 0.002493913
trigram language 0.0024628609999999998
language transcript 0.0024614719999999997
tistical language 0.002447964
poor language 0.002444235
written language 0.0024425989999999997
rich language 0.0024286439999999998
language transcriptions 0.0024254619999999998
language processing 0.002424523
language mod 0.002415524
lingual language 0.0024144429999999996
language modelg 0.002405876
speech translation 0.002237043
language 0.00217719
translation system 0.002053075
text translation 0.00201039
machine translation 0.001979619
word order 0.00196402
word sequence 0.001912399
reference translation 0.001866878
word pairs 0.0018536129999999998
translation accuracy 0.001847437
word error 0.001836805
word pair 0.0018238249999999998
word alignments 0.0018019389999999998
mandarin translation 0.001800852
translation framework 0.0017990789999999999
word strings 0.0017806389999999999
translation tasks 0.001767933
word sequences 0.00175505
monotonic word 0.0017347539999999998
word indexes 0.0017347539999999998
original word 0.0017339109999999999
translation 0.00151589
training data 0.001459596
training set 0.00124982
source speech 0.001247005
text data 0.001234096
syntactic reordering 0.0012309550000000002
phrase order 0.001222412
sequence distribution 0.001218407
phrase number 0.001181819
phrase sequence 0.001170791
parallel data 0.001155733
secutive words 0.001149551
parallel training 0.0011361370000000002
first time 0.001114625
cantonese speech 0.001109955
speech transcription 0.001103699
ing reordering 0.001093947
speech recognition 0.001074788
phrase tables 0.001072491
speech vector 0.0010624179999999999
reordering sequence 0.001058747
reordering wfst 0.001052269
phrase position 0.001049964
different segmentation 0.001045866
speech decoder 0.001043891
training texts 0.001034627
reordering constraints 0.001032053
input speech 0.001029803
models gcl 0.001009618
evaluation method 0.001005652
guage models 0.00100442
parallel corpus 0.001004121
phrase positions 0.001004076
phrase extraction 9.9704E-4
phrase v˜k 9.94477E-4
