Sign Langauge to Speech Conversion

Real-time American Sign Langauge to Speech Conversion using OpenCV and TensorFlow/Keras for hand gesture recognition. Features custom hand tracking, image preprocessing, and gesture classification to translate American Sign Language into text output. Built with accessibility in mind.

⚠️ Note: This project is currently under development.

Features in Development

• Real-time hand gesture detection and tracking
• Image preprocessing pipeline
• Machine learning-based gesture classification
• Support for basic ASL gestures (only letters)

Technical Stack

• Python 3.9+
• OpenCV (Computer Vision)
• TensorFlow/Keras (Machine Learning)
• cvzone (Hand Tracking)
• NumPy (Numerical Processing)

Project Structure

project /
│
├── Application.py 
├── dataCollection.py
├── data 
├── model/             
│   ├── keras_model.h5  
│   └── labels.txt     
├── README.md
└── requirements.txt     

Planned Features

• Text-to-speech functionality
• Expanded gesture vocabulary
• Improved model accuracy
• Support for word prediction
• Support for continuous sentence formation
• GUI interface

Development Status

This project is in active development. Current focus areas:

• Improving hand tracking accuracy
• Expanding the gesture recognition dataset
• Enhancing real-time performance

Requirements

• Python 3.9 or higher
• Webcam or camera device (laptop)
• Key packages: opencv-python, tensorflow, numpy, cvzone, mediapipe

---

More documentation will be added as the project develops.