Akshara 📖

Akshara is an AI-powered reading fluency and literacy improvement platform. It enables users to practice reading passages, receive real-time feedback on pronunciation and fluency, and get personalized coaching through AI-driven acoustic and linguistic analysis.

🚀 Key Features

• Reading & Practice Interface: Interactive reading sessions with audio recording for real-time analysis.
• AI-Driven Speech Analysis:
  • Acoustic Analysis: ML-based classification and speech feature extraction.
  • Linguistic Analysis: Transcription, text alignment, and difference checking.
  • Pre-processing: Voice Activity Detection (VAD) and audio conversion.
• Personalized AI Coaching: Integration with LLMs (Anthropic) to provide tailored feedback and guidance on reading.
• Progress Tracking: Comprehensive history, progress visualization, and result breakdowns.
• Premium UI/UX: A sophisticated "Digital Curator" design philosophy utilizing high-quality typography (Newsreader + Manrope) and glass-morphism.

🛠️ Technology Stack

Frontend

• React 19 (Vite)
• React Router for navigation
• Tailwind CSS & PostCSS for styling

Backend

• Python (Flask)
• MongoDB (Database)
• ML & Audio Processing: PyTorch, Transformers, scikit-learn, ONNX Runtime, librosa, pydub
• AI Integration: Anthropic API

---

💻 Local Development Setup

Prerequisites

• Node.js (v18+)
• Python (3.9+)
• MongoDB (running locally or via Atlas)
• FFmpeg (required for pydub audio processing)

Backend Setup

1. Navigate to the backend directory:

   cd backend

2. Create and activate a virtual environment:

   python -m venv venv
   # On Windows:
   .\venv\Scripts\activate
   # On macOS/Linux:
   source venv/bin/activate

3. Install dependencies:

   pip install -r requirements.txt

4. Set up environment variables (create a .env file):
   • Add your MongoDB URI, JWT Secret, Anthropic API Key, etc.
5. Run the application:

   python run.py

   (Server typically runs on http://localhost:5000)

Frontend Setup

1. Navigate to the frontend directory:

   cd frontend

2. Install dependencies:

   npm install

3. Run the development server:

   npm run dev

   (App typically runs on http://localhost:5173)

📂 Project Structure

E:\Akshara
├── backend/                  # Python Flask server & ML Pipelines
│   ├── app/
│   │   ├── api/              # Route handlers (Auth, Evaluate, Passages)
│   │   ├── database/         # MongoDB models & schemas
│   │   ├── ml_pipeline/      # Core AI logic (Acoustic, Linguistic, Pre-processing, Synthesis)
│   │   └── utils/            # Helper functions
│   ├── run.py                # Backend entry point
│   └── requirements.txt      # Python dependencies
├── frontend/                 # React frontend application
│   ├── src/
│   │   ├── components/       # Reusable UI elements (ProtectedRoute)
│   │   ├── context/          # React Context (AuthContext)
│   │   ├── pages/            # Main application views (Login, Practice, Progress, etc.)
│   │   └── services/         # API integration logic
│   ├── package.json          # Node dependencies
│   ├── tailwind.config.js    # Tailwind setup
│   └── vite.config.js        # Vite bundler config
└── DESIGN.md                 # Design system & philosophy document

📝 License

MIT License