Yes the probabilities need to be position-dependent.<br>Thank you<br>-Sneha<br><br><div class="gmail_quote">On Tue, Mar 16, 2010 at 7:07 PM, Andreas Stolcke <span dir="ltr"><<a href="mailto:stolcke@speech.sri.com">stolcke@speech.sri.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><div><div></div><div class="h5">On 3/15/2010 1:51 AM, Sneha Ann Mathew wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Hi All,<br>
<br>
Is there any option in SRI LM for creating a LM based on i-distant bigrams.This is to include the context of i neighbouring words but considering the bigram combination as follows:<br>
<br>
p(Wn|Wn-4,.....Wn-1)=sum of [Li * p(Wn|Wn-i)] where i varies from 1 to 5 [for my case]<br>
Sum of Li should be 1<br>
</blockquote></div></div>
Are the p(.|.) position-specific, or shared?<br>
Either way, no, this is not implemented in SRILM. But it would be quite straightforward to do so. Here is your chance to contribute!<br>
<br>
Andreas<br>
<br>
<br>
_______________________________________________<br>
SRILM-User site list<br>
<a href="mailto:SRILM-User@speech.sri.com" target="_blank">SRILM-User@speech.sri.com</a><br>
<a href="http://www.speech.sri.com/mailman/listinfo/srilm-user" target="_blank">http://www.speech.sri.com/mailman/listinfo/srilm-user</a><br>
</blockquote></div><br>