Logo Lanfrica

osmernd/Sharhi-hausa-app

Domain:

natural language processingeducation

Record type:

software
Creator:
osm
Host:
# Sharhi πŸ“šπŸš€ > **Juya Ilimi Zuwa Fahimta** (Turning Knowledge into Understanding) ## πŸ’‘ The Problem Millions of students in Northern Nigeria struggle to comprehend complex academic materials because most AI study tools (like NotebookLM or ChatGPT) are English-centric. They lack the cultural nuance and deep language support required for effective learning in **Hausa**. ## πŸ›  The Solution **Sharhi** is the first AI-powered study companion specifically optimized for the Hausa language. It transforms static PDF documents into interactive, localized learning experiences. By leveraging **Google's Gemini 2.5 Flash** and **ElevenLabs**, Sharhi grounds its intelligence in the user's uploaded material but interacts entirely in Hausaβ€”making education accessible, engaging, and native. ## ✨ Key Features * **πŸ“„ Context-Aware Hausa Chat:** Upload any PDF (textbooks, handouts), and chat with it in Hausa. The AI is "grounded" in your document, preventing hallucinations. * **🎧 Audio Overview (Podcast Mode):** Converts dense text into a friendly, engaging Hausa audio summary using **ElevenLabs** (or Google TTS) for on-the-go learning. * **πŸ—‚ AI Flashcards:** Automatically generates study flashcards from your material using **Structured JSON Output**, perfect for active recall. * **πŸ“ Gamified Quizzes:** A dynamic quiz engine that tests your knowledge, tracks your score, and provides instant feedback in Hausa (Pass/Fail). ## πŸ— Tech Stack This project is a Next.js application built for speed and performance: * **Framework:** Next.js 15 (App Router) * **Styling:** Tailwind CSS + Lucide React * **AI Model:** **Gemini 2.5 Flash** (via Google REST API) * **Text-to-Speech:** **ElevenLabs** (Multilingual v2) / Google Cloud TTS * **PDF Processing:** `pdf2json` (Server-side parsing) --- ## πŸ’» Getting Started ### 1. Prerequisites * Node.js 18+ * Google Gemini API Key * ElevenLabs API Key (Optional, for best audio) ### 2. Installation First, clone the repository and install dependencies: `` …

Languages