Python API for reading and querying ARPA format language model.
Usage :
lm = LM("test.arpa")
# Score a sentence
lm.score("HELLO WORLD")
# Get integer word id
wId = lm.get_word_id("HELLO")
# Get the bigram probability of two words
lm.get_bg(lm.get_word_id("HELLO"),lm.get_word_id("WORLD"))
# Get the trigram probability of three words
lm.get_tg(lm.start,lm.get_word_id("HELLO"),lm.end)