This is a csv file which is generated from MatrixCC Dataset available at https://opus.nlpl.eu/results/en&ta/corpus-result-table When downloading the dataset as zip file it contains two seperate files containing tamil and it's equivalent english sentences seperated by "\n" So I merged those to data as one csv for convenience.