Skip to content

This project is to provide spell check help from Urdu to Hindi transliteration.The spelling errors in our case mostly comprises of errors in matras.

Notifications You must be signed in to change notification settings

nausheenfatma/Spell-Check-Using-Bigram-Language-Modelling

Repository files navigation

1)Tokenise Hindi corpus.
python indic_tokenizer.py --i  corpus.txt --o corpus_tokenised.txt --l hin


indic_tokenizer is a sentence tokenizer for Indian Languages. "hin" is the code for the language "Hindi"

2)Run the spell checker (Give an input sentence and the index of the word in consideration) 
python BigramModelSpellCheck.py

The code returns a ranked list of word suggestions (with most probable word on the top)

About

This project is to provide spell check help from Urdu to Hindi transliteration.The spelling errors in our case mostly comprises of errors in matras.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages