Formula Based Regional Dialect Identification of Telugu language Using LDA

Abstract

  Dialect is usually used to signify the language based on the regions or how particularly that is spelled by the local people. Telugu is standard and historical language where we can find four dialects such as Coastal Andhra Slang , Mid Andhra Pradesh slang, Rayalaseema slang , Telangana slang for these Dialects we have created databases such as Andhra (Coastal Andhra Slang + Mid Andhra Pradesh slang), Rayalaseema slang, Telangana slang . There is no standard data base either in speech or text format to identify the regional dialects. This is the main reason to less research in Telugu dialects. In this we created standard database to identify the dialects of Telugu language in text format and we digitalized the data set for pattern recognition, for this we utilized Anu-Script manager to give it a base in the form of formula. We applied to Linear discriminant  analysis pattern recognition algorithm to identify the required pattern which is used in identifying  the  dialect to which  the  word belong to.