word accuracy 0.0026886590000000004
syl word 0.0025198720000000003
word structure 0.0025155710000000003
german word 0.0024519760000000002
probable word 0.0024334580000000003
word figure 0.0024029060000000002
word con 0.0023766860000000002
example word 0.002369302
input word 0.002353477
word forms 0.002322387
word accu 0.002309184
word luftloch 0.0023036420000000003
orthographic word 0.002296489
monosyllabic word 0.002294024
training corpus 0.00214965
syllable model 0.00211689
grammar rules 0.001994185
syllable grammar 0.0019262150000000002
possible syllable 0.001742189
same training 0.001671263
syllable accuracy 0.0016630990000000001
morphological structure 0.001617901
probability model 0.001556017
evaluation corpus 0.0015330069999999999
syllable segmentation 0.0015284530000000001
morphological analysis 0.0015282
pronunciation rules 0.0015254729999999998
training procedure 0.001508793
corpora training 0.0014936519999999998
frequent words 0.001487704
test corpus 0.001467142
known words 0.0014565770000000001
foreign words 0.0014442360000000002
same grammar 0.001442408
compound words 0.0014422620000000001
type training 0.00143783
training iterations 0.001426998
following training 0.0014238249999999999
current model 0.001415969
additional rules 0.00141483
training interations 0.0014089139999999998
grammar models 0.001393211
trained model 0.001379665
training iter 0.001375902
training proce 0.0013735519999999999
training strategy 0.001370837
typenized training 0.001369878
ble model 0.001369035
syllable boundaries 0.0013676060000000002
corpus size 0.001343318
rst syllable 0.001340379
supplementary rules 0.001336327
ability model 0.0013348190000000001
brown corpus 0.001323094
rule probabilities 0.001315138
syllable boundary 0.001313579
news corpus 0.001299567
new grammar 0.001295374
english data 0.001291925
newspaper corpus 0.001291393
syllable parts 0.001288366
sternzeit corpus 0.001261559
test data 0.001242366
start grammar 0.001238269
possible segmentation 0.001218382
words 0.00120055
rst rule 0.001144195
elaborate grammar 0.001142823
grammar fragment 0.001141004
training 0.00112894
subsequent rule 0.00111899
rules 0.0010941
model 0.00109076
learning algorithm 0.0010751939999999998
rule templates 0.001073719
recursive rule 0.001072609
possible analyses 0.001034236
several approaches 0.001027113
syllable 0.00102613
corpus 0.00102071
accuracy values 9.67328E-4
high number 9.672420000000001E-4
other systems 9.668070000000001E-4
several errors 9.503230000000001E-4
main error 9.38129E-4
good results 9.372650000000001E-4
new algorithm 9.369319999999999E-4
syl syl 9.36364E-4
unrestricted text 9.35132E-4
further error 9.34306E-4
tts system 9.321430000000001E-4
several itera 9.25806E-4
several possi 9.25806E-4
coda syl 9.19197E-4
average number 9.188110000000001E-4
maximal accuracy 9.16994E-4
test set 9.1504E-4
coda structure 9.14896E-4
accuracy curve 9.02178E-4
grammar 9.00085E-4
