About the Lab
Sanskrit Computational Research Lab (Sanganaka)
Tagline: सा विद्या या विमुक्तये
“That is knowledge which liberates.”
The Sanskrit Computational Research Lab (Sanganaka) at IIT Kharagpur aims to promote the frontiers of computational linguistics for Sanskrit, which is one of the oldest and structurally richest languages in the world. Led by Prof. Pawan Goyal, the lab discovers interdisciplinary research at the interface of artificial intelligence, classical linguistics, and Indian knowledge systems.
Our goal is to bring Sanskrit to contemporary audiences by creating strong, scalable, and linguistically informed tools and models. Along the way, we hope to revive and protect classical Indian knowledge through the application of state-of-the-art AI methods and low-resource language modeling techniques.
The core research areas in the lab are:
- Sanskrit word segmentation (Sandhi-viccheda)
- Morphological analysis and tagging
- Syntactic dependency parsing
- Prose-to-poetry and poetry-to-prose conversion (Anvaya)
- Compound structure identification (Samāsa)
- Machine translation and multilingual NLP for Sanskrit
- Paninian grammatical frameworks integrated into contemporary NLP pipelines
Our group has created cutting-edge materials like SanskritShala—a neural Sanskrit NLP suite—and DepNeCTI, a novel architecture for detecting nested compounds in ancient texts. These initiatives are part of the development of an extensive digital framework for Sanskrit to facilitate education, scholarship, and cross-linguistic discovery.
The research of extracting knowledge from the lab connects traditional knowledge and contemporary computation, and it provides a model for technological scholarship in the Indic environment. By this work, we attempt to turn Sanskrit from an ancient language of the past into a computationally empowered resource of the future.
