India Showcases 'AI for All' Initiative

- India's national digital education platform, DIKSHA (Digital Infrastructure for Knowledge Sharing), is being integrated with Personalized Adaptive Learning (PAL). This AI-based approach aims to offer individualized learning experiences by, for example, identifying a student's error in a math problem and automatically providing a foundational video to correct the misunderstanding. - The government's BHASHINI initiative is a key enabler for voice-based educational tools, developing translation and speech-to-text models for over 36 Indian languages. This platform aims to create large datasets for Indian languages that can be used to train AI models, a crucial step for developing speech recognition that works for children with diverse accents. - Research on speech recognition for Indian children has shown that commercial APIs can perform poorly, highlighting the need for specialized models. Efforts like AI4Bharat are creating open-source Automatic Speech Recognition (ASR) models for 22 official Indian languages, using thousands of hours of transcribed audio to improve accuracy. - The National Education Policy (NEP) 2020 explicitly calls for integrating AI and computational thinking into the curriculum, starting as early as Grade 3 by the 2026-27 academic year. An expert committee from IIT Madras is developing the curriculum framework, which will focus on concepts and ethical use rather than just technical skills for younger students. - The All India Council for Technical Education (AICTE) declared 2025 the "Year of AI," launching initiatives to integrate applied AI across all technical and higher education programs, not just computer science. This includes faculty development programs to train educators in teaching AI and adapting to AI-driven changes in learning. - A key public-facing component of the strategy is the "AI For All" self-paced learning program, a 4-hour online course developed with Intel and CBSE to demystify AI for students, parents, and professionals. The program is available in 11 vernacular languages and is designed to be accessible, with content compatible with talkback applications for the visually impaired. - To address learning disabilities, the AI-powered "Voice Fusion Learn" platform is being developed for early diagnosis and personalized support for students with Dyslexia, Dysgraphia, and Dyscalculia. It uses a multimodal diagnostic engine that combines voice analysis, handwriting recognition, and typing behavior to create tailored, voice-guided learning interventions.