word accuracy 0.002505739
training data 0.00243259
training words 0.00242722
word level 0.00237133
word length 0.0022733090000000003
entire word 0.002252887
italian word 0.002233333
english letter 0.002230679
word accuracies 0.002222288
letter context 0.00220207
english data 0.002198719
unseen word 0.002194908
absolute word 0.002192159
word markers 0.002192143
word marker 0.00219007
word scianchi 0.00219007
erage word 0.00219007
other words 0.002083489
letter clustering 0.0020426669999999997
letter class 0.0020205609999999997
letter sequence 0.002006785
clustering words 0.0020053370000000003
letter classes 0.001910845
letter level 0.00180949
test words 0.001798214
letter distance 0.001756382
available data 0.001750607
phoneme features 0.001739106
following letter 0.001725353
string letter 0.0017215009999999998
letter symbols 0.001718687
ipa letter 0.001712635
small data 0.001703997
data samples 0.0016987740000000001
learning algorithm 0.001686503
such features 0.001685767
letter bit 0.001678866
candidate words 0.001676839
training set 0.0016714380000000001
letter hierarchy 0.001668554
letter counts 0.001663611
focus letter 0.0016595479999999998
development data 0.001646976
pronalsyl letter 0.001641239
ineligible letter 0.0016311429999999998
ual letter 0.001630079
surrounding letter 0.001629616
cmu data 0.001629423
ery letter 0.00162865
distant letter 0.00162865
unlabelled data 0.001625237
annotated data 0.001614517
pool data 0.001610434
labelled data 0.001607252
labelled words 0.001601882
frequent words 0.001600979
unseen words 0.001595738
tated words 0.001592665
class features 0.00159253
notated words 0.001591036
learning method 0.00148569
available training 0.001479017
learning set 0.0014779110000000002
clustering algorithm 0.001458147
small training 0.0014324070000000001
phoneme sequence 0.0014058220000000001
supervised learning 0.001401828
other clustering 0.001395386
training sets 0.001390965
letter 0.00138405
initial training 0.0013802
tree learning 0.001368954
english figure 0.001363883
alignment method 0.001348468
words 0.00134672
labelled training 0.001335662
training exam 0.001326415
adequate training 0.001323547
english letters 0.001311059
phonetic alignment 0.0013074710000000002
active learning 0.001283064
learning techniques 0.001282609
context letters 0.00128245
learning strategy 0.001279556
tree model 0.001278851
similar alignment 0.001257146
learning technique 0.001253192
other approaches 0.001201708
entire learning 0.00119397
learning setting 0.001173191
learning curve 0.001170605
single phoneme 0.00116396
tive learning 0.001154264
system baseline 0.001147197
baseline system 0.001147197
context length 0.001145439
sive learning 0.0011395210000000001
learning curves 0.0011325369999999999
standard approach 0.001132493
english dataset 0.0011320430000000001
