word model 0.00420165
model results 0.003190904
large model 0.003126196
model parameters 0.002982555
generative model 0.002971244
model space 0.002949264
length model 0.002929814
stochastic model 0.002911387
direct model 0.0028954000000000002
transducer model 0.002870583
ory model 0.002811962
mci model 0.002805836
graphical model 0.0027948260000000003
memory model 0.002788614
model memory 0.002788614
basic model 0.002786639
bahl model 0.0027830990000000002
model memoryless 0.002782071
based model 0.002781362
unrolling model 0.002778066
explicit model 0.0027779610000000002
model 0.00255382
word sequence 0.0021401270000000003
next word 0.0019836050000000003
word pronunciations 0.0019306940000000002
word resp 0.001869501
training words 0.001601456
other models 0.00155759
training data 0.0014391500000000002
target string 0.001403678
different edit 0.001371673
string edit 0.001364197
models performance 0.001357933
models modeling 0.001308149
context models 0.001305472
distance models 0.001292759
source string 0.001262693
position models 0.00125595
several models 0.001247111
previous models 0.001224494
length models 0.001222169
stochastic models 0.001203742
string pair 0.001201699
corresponding words 0.0011978219999999999
test words 0.001194645
same results 0.0011935280000000001
language processing 0.001191007
dbn models 0.001184625
other pronunciation 0.001183679
models framework 0.001176634
string context 0.0011674139999999999
natural language 0.001162741
target context 0.001154858
edit sequence 0.001148377
different speech 0.001147061
probability distribution 0.00114423
ing string 0.001135191
recognition models 0.00113022
modeling edit 0.0011180539999999998
string position 0.001117892
markov models 0.00111644
machine translation 0.001112479
target position 0.0011053360000000002
cal models 0.001104427
edit distance 0.001102664
conditional probability 0.00110264
string pairs 0.001089067
graphical models 0.001087181
error rate 0.001081191
memory models 0.001080969
hidden edit 0.0010735179999999999
models perspective 0.001067519
small edit 0.001055521
test set 0.001039085
other states 0.001037118
different strings 0.001036244
different types 0.0010316470000000001
edit operations 0.001031511
other types 0.001027469
learning edit 0.001027053
chine translation 0.001025076
such example 0.001023038
target strings 0.001016212
probability table 0.001014723
source context 0.001013873
cost function 0.001011934
results table 0.001011097
different pronunciations 9.984570000000001E-4
text dependencies 9.961100000000001E-4
joint probability 9.93879E-4
target symbol 9.89682E-4
other cases 9.71697E-4
target symbols 9.66899E-4
key feature 9.651290000000001E-4
different parent 9.64826E-4
source position 9.64351E-4
training hmms 9.61033E-4
conditional probabilities 9.606459999999999E-4
other problems 9.56801E-4
parent set 9.554769999999999E-4
