• Bhasha Audibles : http://cvit.iiit.ac.in/bhashaaudible/
  • This innovative application transforms scanned books into high-quality audiobooks in MP3 format, making knowledge and literature more accessible to diverse audiences. With a focus on clarity and immersive listening, it provides smooth narration and excellent audio fidelity. Currently, the platform supports three major Indian languages—Hindi, Telugu, English and Punjabi—ensuring that users can enjoy content in their preferred language. Ideal for visually impaired individuals, language learners, and anyone on the go, this tool bridges the gap between traditional print and modern audio technology.

  • Punjab Vidhan Sabha Proceedings Search Engine : http://101.53.145.85:6022/vidhan-search/
  • This application enables the digitization and searchability of Punjab Vidhan Sabha proceedings, significantly enhancing accessibility and transparency. By converting historical and current records into a searchable digital format, it allows users to efficiently access legislative discussions, decisions, and archives, fostering greater public engagement and accountability.

  • IndicOCR : http://ilocr.iiit.ac.in/indicocr/
  • This application efficiently extracts accurate and reliable text from scanned documents, offering robust support for over 22 Indian languages. It ensures high precision in text recognition, making it an essential tool for multilingual document processing and analysis.

  • IndicPhotoOCR : https://github.com/Bhashini-IITJ/BharatSceneTextDataset
  • This tool is designed for seamless usability, enabling word recognition from an image with just three lines of code. Its simplicity, coupled with the power of containerization, ensures quick deployment across various environments, opening up limitless possibilities. Users can effortlessly integrate it into their workflows, even with limited computational resources. The ease of use, combined with the flexibility of deployment, makes IndicPhotoOCR an invaluable resource for anyone looking to leverage scene text recognition in their applications.