Structured perceptron based transliteration system for cross transliterations among 16 Indian languages including English and Urdu.
Universal Dependency Treebank for Hindi-English Code Switching.
Language identification and normalisation in code switching data tailored with a three-step decoding process.
Neural Stacking Dependency Parsers for monolingual, multilingual and code switching texts.
Python library for UTF to WX conversion and vice-versa for Indian languages.
Tokenizer for world’s most spoken languages and social media texts like Facebook, Twitter etc.