evidoria

← Back to browse

Good practice Imported

Bhashini and NCERT's AI Pipeline Translates India's National Textbooks Into 22 Languages

India · New Delhi · See the India profile · See the New Delhi profile

Evidence: Descriptive / self-reported Top 81% 33/100 · Ask Evidence Copilot about this practice

India's government AI translation platform Bhashini and NCERT's CIET are using generative AI to translate national curriculum textbooks and materials across 22 scheduled Indian languages.

22 languages
Scheduled Indian languages covered by the translation pipeline
Bhashini and NCERT's AI Pipeline Translates India's National Textbooks Into 22 Languages

Details

Promoter
Ministry of Electronics and Information Technology (Bhashini/NLTM) / NCERT's Central Institute of Educational Technology
Period
2022-2025
Keywords
generative AI translation, curriculum content, multilingual education, textbook authoring, government platform

Context

Bhashini is an AI-driven language-technology platform launched by India's Ministry of Electronics and Information Technology (MeitY) in July 2022 under the National Language Translation Mission, offering machine translation, speech recognition and text-to-speech across India's 22 scheduled languages. NCERT's Central Institute of Educational Technology (CIET) runs a dedicated Translation Cell applying AI-assisted translation to national textbooks, teacher-training materials and policy documents.

Objectives

The stated policy driver is India's National Education Policy 2020, which calls for instruction in a child's home or regional language through at least Grade 5, and ideally Grade 8 — a goal that is logistically very difficult without machine-assisted translation given India's linguistic diversity.

Activities

NCERT-CIET's Translation Cell renders national textbooks, teacher-training materials and policy documents out of Hindi/English into regional languages such as Tamil, Bengali, Marathi, Odia and Assamese. Separately, India's higher-education regulators AICTE and UGC use a related tool, Anuvaadini, to machine-translate undergraduate and postgraduate technical textbooks, published on the government's e-KUMBH portal.

Conclusions

Available public sources describe the scale of the translation infrastructure and its policy rationale, but do not provide an independent, learning-outcome evaluation of AI-translated textbook content; machine translation of technical and culturally specific content into low-resource Indian languages is acknowledged in NLTM's own materials as an ongoing quality challenge requiring human review.

Implementation

Indicative cost
High (€500k–€5M) — National-scale AI infrastructure spanning multiple ministries and agencies (MeitY, NCERT, AICTE, UGC); no public budget figures found.
Time to results
Long (> 3 years) — Bhashini platform launched July 2022 under NLTM; NCERT/CIET translation activity and AICTE/UGC's Anuvaadini tool operate on an ongoing basis through at least 2025.
Staffing & skills
Ministry of Electronics and Information Technology (MeitY) / National Language Translation Mission technical team (Bhashini platform), NCERT's Central Institute of Educational Technology (CIET) Translation Cell, AICTE and UGC, operating the related Anuvaadini tool for higher-education technical textbooks

Conditions for success

  • A national policy mandate (NEP 2020) creating sustained demand for mother-tongue instructional materials
  • Bhashini's AI models and APIs available as pluggable public infrastructure for other government bodies and ed-tech providers
  • A human review layer for technical and culturally-specific content

Common failure modes

  • Machine translation quality challenges for technical and culturally specific content in low-resource Indian languages, acknowledged by NLTM's own materials
  • No independent evaluation exists of translated-textbook accuracy or its effect on student learning outcomes

Where it fits

Governance type
national government, multi-agency
Scale
national (22 languages, multiple ministries/agencies)
Income level
lower-middle-income

Commonly funded by

National / regional programmes

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful