same word 0.003794996
vocabulary word 0.003724836
word adaptation 0.003652732
word frequency 0.003652089
consistent word 0.003642102
tagalog word 0.00363985
length word 0.003639059
next word 0.003604397
word sequence 0.003602469
word occurrence 0.0035792289999999997
word burstiness 0.003571737
word tokens 0.003571706
particular word 0.003569452
individual word 0.003559062
word error 0.0035574359999999998
word types 0.0035539869999999998
word count 0.003527471
meaningful word 0.003526231
word repetition 0.003522404
word detections 0.00351875
word occurrences 0.003518056
word repetitions 0.003515914
entire word 0.003513542
word repe 0.003511236
word frequen 0.003511236
putative word 0.003511236
word bursts 0.003511236
word scope 0.003511236
language model 0.0029185699999999997
term detection 0.002239883
topic model 0.002197146
model information 0.002138562
language models 0.002112245
search term 0.002063305
term development 0.0019898159999999997
repeated term 0.0019517439999999998
same language 0.001930926
spoken term 0.0019211719999999999
cts term 0.001920675
term detec 0.001916337
term detections 0.00191601
tual term 0.0019123019999999998
top term 0.0019103269999999999
poisson model 0.0019053639999999999
frequency words 0.001817089
acoustic model 0.001801261
model perplexity 0.00178989
similar language 0.0017848879999999999
cache language 0.0017834589999999998
mixture model 0.001755865
guage model 0.0017536519999999999
many language 0.001749121
model configurations 0.00173863
frequent words 0.001723398
repeated words 0.0017194839999999999
content words 0.001704595
absolute language 0.001694814
quency words 0.001685996
turkish language 0.001685269
training data 0.00168015
gram language 0.00166522
language mod 0.001665175
language condition 0.0016508529999999999
adaptive language 0.001647401
model 0.00150491
words 0.00144273
english data 0.00143851
language 0.00141366
other information 0.001404649
topic context 0.001396311
topic models 0.001390821
context information 0.001337727
topic information 0.001325888
training corpus 0.001280338
training vocabulary 0.00126433
tagalog data 0.001225046
document context 0.001211854
babel training 0.001211689
training speech 0.001197936
tagalog training 0.001179344
development data 0.001177752
such information 0.001135449
latent topic 0.0011266330000000001
dev data 0.001102999
training transcripts 0.001101715
different interpolation 0.001099282
asr training 0.001093658
additional terms 0.0010845310000000001
detection scores 0.001062712
detection score 0.0010606930000000001
training tran 0.001060353
training condition 0.001054417
training conditions 0.001052623
latent semantic 0.001045398
target corpus 0.001044348
topic contexts 0.001036435
unigram probability 0.001029094
same document 0.001025045
ment context 0.001020803
adaptation probability 0.001019875
