hi
I made a bigram language model on a small document with both HTK toolkit and SRILM toolkit.
but surprisingly their accuracy in HTK based ASR are so different. I mean when I use  SRILM bigram the accuracy is 10 percent lower than the one I build with HTK.
I am confused I would appreciate if someone help me where is my mistake?

Here is my command for build bigram in SRILM:
    ngram-count -text sentences.txt -order 2 -wbdiscount 1 -wbdiscount 2 -lm bigram.txt
    sentences.txt has 405 sentences.

I build my acoustic model based on left to right HMMs with 16 Gaussian mixture for triphones using HTK.


Thanks

--- (Edited on 6/20/2011 5:06 am [GMT-0500] by m_ice) ---