TLDR: Researchers at IIT Jodhpur have developed an indigenous, open-source AI model named Indic Photo Optical Character Recognition (OCR) capable of reading and translating text from signboards and street signs in 13 major Indian languages. This initiative, part of the central government’s ‘Bashini’ program, aims to break down linguistic barriers, enhance digital accessibility, and preserve India’s diverse cultural heritage.
Jodhpur, Rajasthan – The Indian Institute of Technology Jodhpur has announced a significant breakthrough in artificial intelligence with the development of an indigenous AI model designed to read and translate signboards and street signs across 13 prominent Indian languages. This innovative technology, named Indic Photo Optical Character Recognition (OCR), was created by the Department of Computer Science and Engineering, specifically by the Vision Language and Learning Group (VL2G) team, under the leadership of Professor Anand Mishra.
The Indic Photo OCR model is an open-source solution, developed as part of the central government’s ‘Bashini’ initiative, which focuses on enhancing language accessibility and addressing linguistic diversity within the country. Professor Anand Mishra stated, ‘We have developed Indic Photo OCR, which is an open-source model. It is a completely indigenous model. It has been created to read and translate Indian languages in signboards and street signs.’
The model boasts the capability to accurately interpret scene text in 13 major Indian languages, including Hindi, Marathi, Gujarati, Punjabi, Bengali, Assamese, Odia, Manipuri, Malayalam, Kannada, Tamil, Telugu, and Urdu. This comprehensive linguistic support is tailored to India’s rich linguistic landscape.
The primary objectives of this AI tool extend beyond mere translation. It aims to dismantle language barriers, thereby enhancing digital access for millions of Indians, making the internet more inclusive, and empowering communities nationwide. Professor Mishra emphasized that the technology is designed to be freely accessible to everyone, contributing meaningfully to the preservation of India’s cultural heritage by advancing recognition and support for its diverse languages.
Also Read:
- Karnataka Unveils Ambitious AI Mission and Policy Framework
- OpenAI Reveals Half of Indian ChatGPT Users are Students Under 24, Announces Major Education Initiatives
Furthermore, this development aligns with a broader vision to establish India as a global leader in artificial intelligence, not only by driving innovation but also by delivering solutions uniquely suited to the nation’s social and cultural context. IIT Jodhpur is also actively involved in other AI-driven preservation efforts, collaborating with the Indira Gandhi National Centre for the Arts (IGNCA) and TIH-iHUB Drishti. This partnership focuses on digitally preserving and restoring ancient manuscripts written in Sanskrit, Pali, Telugu, and other classical languages, utilizing deep learning and advanced computer vision techniques to repair damaged texts for long-term digital archiving.
