AI4Bharat is a center at IIT Madras with the mission of bringing parity in AI technologies for Indian languages with respect to English with open-source contributions in datasets, neural models, tools for data collections, and reference applications. This mission is cross-cutting in terms of (a) languages supporting all 22 constitutionally recognized languages in India, and (b) tasks supporting machine translation, speech recognition, speech synthesis, character recognition, and language understanding.AI4Bharat has received a grant of 47 crore INR (6 million USD) from the Government of India as part of the Bhashini project which is India’s most ambitious language AI project till date. As part of this project (March 2022 – March 2024), AI4Bharat will be creating open-sourced datasets across languages and tasks often improving on existing datasets by orders of magnitude. AI4Bharat will also be building an open-source framework for collaboration on language data collection and annotation. In addition, AI4Bharat has also received a funding of 32 crore INR (4.2 million USD) from Nandan Nilekani who has been instrumental in several population-scale efforts such as Aadhaar, UPI, and ONDC. As part of this project, AI4Bharat is committed to building real-world applications that can be deployed across government, NGOs, and startups.In summary, with its research strength, deep collaboration with the government, and intent to build real-world applications, AI4Bharat is poised to provide leadership in the development of AI technologies for Indian languages over the next few years. With a principled stance on open-sourcing all output with the most liberal licensing terms, AI4Bharat is open to collaborations with research groups, data partners, and cloud/hardware companies offering compute.
Want to print your doc? This is not the way.
Try clicking the ⋯ next to your doc name or using a keyboard shortcut (