Scan textbook pages or notes using OCR. EduLearn AI extracts text, explains concepts, and generates offline study notes instantly.
Explore personalized career paths, eligible courses, colleges, and entrance exams after 10th and 12th to make informed education decisions.
EduLearn AI provides offline, syllabus-based answers with simple explanations, and one-tap options to listen, copy, or save notes.
Offline AI Tutor welcomes students with personalized, multilingual, curriculum-based learning and suggested questions.
Students can learn through voice by speaking in Tamil, English and receiving offline AI-powered speech recognition and pronunciation support
Offline Study Notes help students store important AI explanations, organize subject-wise notes, and revise anytime, anywhere.
Home Dashboard – Access AI Tutor, Voice Learning, Camera OCR, Quizzes, Notes, Career Guidance, and learning progress from one screen.
The Progress Dashboard tracks learning streaks and study performance, helping students monitor their academic progress and milestone badges.

Inspiration

Millions of students in rural and semi-urban areas lack reliable internet access, making it difficult to use existing AI learning platforms. Most educational AI assistants depend on cloud services, require subscriptions, and offer limited support for regional languages. We wanted to build an AI learning companion that works completely offline, supports multiple languages, and provides every student with equal access to quality education and career guidance.

What it does

EchoLearn is a fully offline multilingual AI learning assistant that helps students learn anytime, anywhere without an internet connection.

Students can ask questions by typing, speaking, or scanning textbook pages and handwritten notes using the camera. The app retrieves relevant curriculum content from a local knowledge base using Retrieval-Augmented Generation (RAG) and generates accurate, syllabus-based answers through an on-device language model. It also provides offline voice responses in the user's preferred language and offers career guidance by recommending suitable degree courses and college options based on academic stream and cutoff marks.

How we built it

We built the mobile application using Flutter for a cross-platform experience. Whisper enables offline multilingual speech recognition, while Tesseract OCR extracts text from textbook pages and handwritten notes. A lightweight quantized LLM running through llama.cpp performs on-device reasoning. A local vector database (FAISS/ChromaDB) powers the RAG pipeline, ensuring curriculum-grounded responses. Piper TTS provides offline multilingual voice output, allowing the entire system to function without internet connectivity.

Challenges we ran into

Building a completely offline AI system was our biggest challenge. We had to optimize model size and memory usage so it could run efficiently on mobile devices while maintaining good response quality. Integrating OCR, speech recognition, RAG, and multilingual support into a single seamless workflow also required careful optimization and testing

Accomplishments that we're proud of

Built a fully offline AI learning assistant with no internet dependency.
Integrated text, voice, and image-based question input.
Implemented curriculum-grounded RAG to reduce AI hallucinations.
Added multilingual support for over 20 languages with offline voice responses.
Included personalized career and college guidance for students.

What we learned

Through this project, we gained hands-on experience with offline AI deployment, Retrieval-Augmented Generation (RAG), vector databases, speech recognition, OCR, multilingual NLP, Flutter development, and mobile AI optimization. We also learned the importance of designing AI solutions that are accessible, inclusive, and practical for students in low-connectivity environments.

What's next for EchoLearn - Offline Multilingual AI Learning Assistant

We plan to expand the curriculum database to support more education boards and competitive exams, improve answer quality with better offline models, introduce personalized learning plans and progress tracking, add teacher and parent dashboards, and optimize the application to run efficiently on low-end smartphones. Our long-term vision is to make high-quality AI-powered education accessible to every student, regardless of language or internet availability.