colab1sandbox/kikuyu-project

# 🌍 Kikuyu-English Translation Collection Platform A mobile-first, community-powered web platform for collecting Kikuyu-English translation data. This tool guides users through structured prompts generated by a language model (LLM) and stores their translations for future use in training a machine translation model. ## 🎯 Project Goals - **Preserve and promote the Kikuyu language** through digital documentation - **Collect high-quality, diverse translation pairs** from native speakers - **Use structured prompts** to avoid repetition, abuse, and low-quality input - **Empower human reviewers** to moderate and validate submissions - **Build a dataset** suitable for training a Transformer-based translation model ## 🛠️ Tech Stack ### Frontend - **HTML5 + CSS3** - **Tailwind CSS** (responsive, mobile-first design) - **JavaScript** (vanilla JS for interactivity) ### Backend - **Flask (Python)** – lightweight and scalable - **SQLite** – simple local database for storing prompts and translations - **Jinja2** – Flask's templating engine for rendering pages - **SQLAlchemy** – ORM for database operations - **WTForms** – form validation ### LLM Integration - **OpenRouter API** – for generating culturally relevant English prompts - **Local JSON caching** – efficient prompt storage and delivery - **Automatic refill system** – maintains prompt availability ## 🚀 Quick Start ### Prerequisites - Python 3.8 or higher - OpenRouter API key (get one at openrouter.ai) ### Installation 1. **Clone the repository** ```bash git clone cd kikuyu ``` 2. **Create virtual environment** ```bash python3 -m venv venv source venv/bin/activate # On Windows: venv\Scripts\activate ``` 3. **Install dependencies** ```bash pip install -r requirements.txt ``` 4. **Configure environment variables** ```bash cp .env.example .env # Edit .env with your configuration ``` 5. **Run the application** ```bash python run.py ``` 6. **Open in browser** Visit `localhost` to access the platform …