A South-African Sign Language Interpreter Web APP, developed with Python as a Third Year BComp Project
# Echo-Me
A **South African Sign Language Interpreter** desktop application developed with Python. This AI-powered application uses computer vision and deep learning to translate sign language gestures into speech in real-time.
## 🌟 Features
- **Real-time Sign Language Recognition** - Uses MediaPipe and TensorFlow for accurate gesture detection
- **Text-to-Speech Conversion** - Converts recognized signs to spoken words with multiple voice options
- **Speech-to-Text** - Voice recognition with microphone input device selection
- **Multi-language Translation** - Translate text before speech synthesis
- **Modern GUI** - Clean, responsive interface built with PySide6
- **User Preferences** - Customizable themes, voice settings, and user accounts
- **Live TTS Toggle** - Enable/disable instant speech output for detected signs
- **Audio Export** - Download generated speech as audio files
## 🎯 Supported Signs
The application recognizes various South African Sign Language gestures including:
**Letters**: a, b, c, d, e, f, g, h, i, k
**Words**: you, me, my, hello, like, apple, orange, pear, name, j, z
**Special**: None (no gesture detected)
## 🛠️ Technology Stack
- **Computer Vision**: MediaPipe, OpenCV
- **Machine Learning**: TensorFlow, Keras, NumPy
- **GUI Framework**: PySide6 (Qt)
- **Speech Processing**: SpeechRecognition, PyAudio, gTTS, pyttsx3
- **Translation**: deep-translator
- **Audio**: pygame
- **Security**: bcrypt, cryptography
## 📋 Requirements
Before installation, ensure you have:
- **Python 3.11**
- **pip** (Python package manager)
- **Git** (optional, for version control)
- **CUDA 11.8+** (optional, for GPU acceleration)
- **Webcam** (for sign language detection)
- **Microphone** (for speech-to-text functionality)
## 🚀 Quick Start
### 1. Clone the Repository
```bash
git clone
github.com
cd Echo-Me
```
### 2. See SETUP_GUIDE.md
GOT TO SETUP GUIDE
## 📁 Project Structure
```
Echo-Me/
├── assets/ …