Prekshagp/A-Neural-Approach-to-Speech-to-Text-Translation-of-Low-Resource-Lamguage

# Konkani Vakya 🌍 **Konkani Vakya** is a modern, AI-powered web application designed to bridge the gap between spoken Konkani and written English. Using advanced linguistic AI patterns, it transcribes Konkani speech and provides accurate English translations, empowering users to learn and preserve the Konkani language. --- ## ✨ Key Features - 🎙️ **Simultaneous Transcription & Translation**: Speak in Konkani and watch as the app transcribes your words into Devanagari script and translates them into English in real-time. - 🗂️ **Personal Flashcards**: Save your translations as flashcards to practice and build your vocabulary. - 🌓 **Modern & Responsive UI**: A premium user interface built with Radix UI and Tailwind CSS, featuring a beautiful dark-mode inspired aesthetic. - 🤖 **AI-Driven Pipeline**: Powered by Genkit and Google Gemini, the app uses a sophisticated multi-step pipeline for language identification, grammatical correction, and transliteration. --- ## 🚀 Tech Stack - **Framework**: Next.js 14 (App Router) - **AI Framework**: Genkit - **Database & Auth**: Firebase (Firestore & Auth) - **Styling**: Tailwind CSS - **Components**: Radix UI & Lucide React - **Validation**: Zod --- ## 🛠️ Getting Started ### Prerequisites - Node.js 18+ - A Google Cloud Project with Gemini API access - A Firebase project for Authentication and Firestore ### Installation 1. **Clone the repository**: ```bash git clone github.com cd konkani-vakya ``` 2. **Install dependencies**: ```bash npm install ``` 3. **Set up environment variables**: Create a `.env.local` file in the root directory and add your Firebase and Genkit credentials: ```env GOOGLE_GENAI_API_KEY=your_gemini_api_key NEXT_PUBLIC_FIREBASE_API_KEY=your_key NEXT_PUBLIC_FIREBASE_AUTH_DOMAIN=your_project.firebaseapp.com NEXT_PUBLIC_FIREBASE_PROJECT_ID=your_project_id # ... other firebase vars ``` 4. **Run the development server**: ```bash npm run dev ``` --- ## 🧠 How …