The folder should contain:
1	unig.txt	~ unigram dictionary file
2	big.txt		~ bigram  dictionary file
3	alphabet	~ hindi alphabet file
4	hi_phonet.dat	- hindi phonetic data file
5	hi-spell1.txt	- comobined spell checkers file
6	hi-spell2.txt	- continous text test file


The unig.txt and big.txt files are not provided for they are huge!

They can be generated using corpus.txt (the combined hindi corpus) by the following python codes:
	clean.py  which creates clean.txt   using corpus.txt and rm.txt
	list.py	  which creates words.txt   using clean.txt
	bigram.py which creates bigrams.txt using clean.txt
	bgfreq.py which creates bgfreq.txt  using bigrams.txt

finally	mkdict.py which creates big.txt and unig.txt using words.txt and bgfreq.txt



Unigram Spell Checker:
	spell1.py needs file 1,3,4,5
Continous Spell Checker:
	spell2.py needs file 1,2,3,4,6
