Abstract
Quantum Natural Language Processing (QNLP) is taking huge leaps in solvingthe shortcomings of classical Natural Language Processing (NLP) techniques andmoving towards a more "Explainable" NLP system. The current literature aroundQNLP focuses primarily on implementing QNLP techniques in sentences in theEnglish language. In this paper, we propose to enable the QNLP approach toHINDI, which is the third most spoken language in South Asia. We present theprocess of building the parameterized quantum circuits required to undertakeQNLP on Hindi sentences. We use the pregroup representation of Hindi and theDisCoCat framework to draw sentence diagrams. Later, we translate thesediagrams to Parameterised Quantum Circuits based on Instantaneous QuantumPolynomial (IQP) style ansatz. Using these parameterized quantum circuitsallows one to train grammar and topic-aware sentence classifiers for the HindiLanguage.