Echo-Sign is a real-time American Sign Language (ASL) interpreter that leverages high-speed computer vision and local multimodal AI to bridge the communication gap. By transforming 3D hand landmarks into natural English text and speech, Echo-Sign serves as a digital bridge for accessibility.

• Real-Time Translation: Uses MediaPipe to track 21 hand landmarks at 30+ FPS.
• Hybrid AI Logic: Combines a Python-based physics engine (Math Layer) with Llama 3.2 (Ollama) to interpret complex gestures with high accuracy.
• Reverse Translation (Text-to-ASL): Type English words to see the corresponding ASL signs (GIFs) or fingerspelling fallbacks.
• Voice Output: Integrated Text-to-Speech (TTS) for hands-free communication.
• Privacy First: The entire translation engine runs locally on your machine using Ollama.

• The Eye: MediaPipe & OpenCV (Hand Tracking)
• The Brain: Llama 3.2 via Ollama (Contextual Sign Interpretation)
• The Voice: pyttsx3 (Text-to-Speech)
• The Interface: Streamlit (Web Dashboard)

1. Prerequisites

   • Install Ollama
   • Pull the model: ollama run llama3.2

2. Clone & Install

   git clone <your-repo-link>
   cd Echo-Sign
   python3.11 -m pip install -r requirements.txt
   
   python3.11 -m streamlit run app.py