raju_rr - 2005-01-14

I created a new lm using CMU stat tool kit. First, there is UNK inserted in the 1-gram list, which I dont' know where it came from. Anyway...

My main problem is occuring when I run the hellongram with this new lm. I changed the config file to incorporate this lm. When I run this, program stops while allocating the decoder. Specifically, the decoder seems to be allocating lextreelinguist, which calls the first node in hmmtree. For some reason that is null. I cannot figure out why. Here is the java error message:
Exception in thread "main" java.lang.NullPointerException
at edu.cmu.sphinx.linguist.lextree.LexTreeLinguist.getInitialSearchState(LexTreeLinguist.java:464)
at edu.cmu.sphinx.linguist.lextree.LexTreeLinguist.compileGrammar(LexTreeLinguist.java:490)
at edu.cmu.sphinx.linguist.lextree.LexTreeLinguist.allocate(LexTreeLinguist.java:407)
at edu.cmu.sphinx.decoder.search.WordPruningBreadthFirstSearchManager.allocate(WordPruningBreadthFirstSearchManager.java:323)
at edu.cmu.sphinx.decoder.Decoder.allocate(Decoder.java:109)
at edu.cmu.sphinx.recognizer.Recognizer.allocate(Recognizer.java:182)
at demo.sphinx.hellongram.HelloNGram.main(HelloNGram.java:69)

When I add a new lm model, do I have to do anything else? I did not change the dictionary or the acoustic model. I've struggled fixing this for several days. Thanks.

rr