--0-103997848-1056390344=:78586
Content-Type: text/plain; charset=us-ascii
Hi,
I tried to build class based LMs in the following way:
step-1: ngram-class -text test.in -numclasses 100 -class-counts text.cnt -classes text.cls -save 100
step-2: ngram-count -read text.cnt -memuse -kndiscount -kndiscount1 -kndiscount2 -lm text.srilm.gz
I found that the class count output "text.cnt" from step-1 is only bigram-counts. Thus the final class-LM text.srilm.gz is also a bigram one.
Could anyone tell me if I am using the toolkit correctly? How to build a trigram class-based LM? Also are there any published paper/document that I can look up for detail information?
Many thanks,
-June
---------------------------------
Do you Yahoo!?
SBC Yahoo! DSL - Now only $29.95 per month!
--0-103997848-1056390344=:78586
Content-Type: text/html; charset=us-ascii
<DIV>Hi,</DIV>
<DIV> </DIV>
<DIV> I tried to build class based LMs in the following way:</DIV>
<DIV> </DIV>
<DIV> step-1: ngram-class -text test.in -numclasses 100 -class-counts text.cnt -classes text.cls -save 100</DIV>
<DIV> </DIV>
<DIV> step-2: ngram-count -read text.cnt -memuse -kndiscount -kndiscount1 -kndiscount2 -lm text.srilm.gz</DIV>
<DIV> </DIV>
<DIV> I found that the class count output "text.cnt" from step-1 is only bigram-counts. Thus the final class-LM text.srilm.gz is also a bigram one. </DIV>
<DIV> </DIV>
<DIV> Could anyone tell me if I am using the toolkit correctly? How to build a trigram class-based LM? Also are there any published paper/document that I can look up for detail information? </DIV>
<DIV> </DIV>
<DIV> Many thanks,</DIV>
<DIV> </DIV>
<DIV>-June</DIV><p><hr SIZE=1>
Do you Yahoo!?<br>
<a href="http://pa.yahoo.com/*http://rd.yahoo.com/evt=1207/*http://promo.yahoo.com/sbc/">SBC Yahoo! DSL</a> - Now only $29.95 per month!
--0-103997848-1056390344=:78586--
Click here to go to the SRILM home page.