🧠 Sign Language Translator Using Computer Vision & Machine Learning

This project is a real-time Sign Language Translator that uses a webcam to recognize hand gestures and convert them into text and speech. It utilizes Computer Vision, Machine Learning, and Deep Learning techniques to interpret sign language effectively.

---

💡 Features

• Real-time gesture recognition using webcam input
• Translates hand gestures into corresponding text
• Converts translated text into speech
• Preprocessing and landmark extraction using MediaPipe
• Model training using scikit-learn and Keras
• Modular code for easy extension

---

🛠️ Project Modules

• image_to_csv.py – Capture hand landmarks and export to CSV
• data_preprocessing.py – Normalize and preprocess dataset
• model_training.py – Train gesture recognition model
• predict_gesture.py – Load model and predict gestures
• real_time_gesture.py – Real-time webcam detection and translation

---

✅ Requirements to Run

Ensure the following are installed:

• Python 3.9.12
• Libraries:
  • opencv-python
  • mediapipe
  • numpy
  • pandas
  • scikit-learn
  • keras
  • pyttsx3 (for speech output)

You can install dependencies with:

pip install opencv-python mediapipe numpy pandas scikit-learn keras pyttsx3

---

🚀 How to Run

1. Capture landmarks
   Run image_to_csv.py to collect gesture data (if required)

2. Preprocess data
   Run data_preprocessing.py to prepare data for training

3. Train the model
   Run model_training.py to train and save your model

4. Predict single image
   Use predict_gesture.py to predict a gesture from static image/landmarks

5. Real-time recognition
   Launch real_time_gesture.py to start webcam detection and translation

---

⚠️ Changes Needed In Order To Run The Code

• In image_to_csv.py, change the path to point to your dataset image directory before running the script.
  • Example:
  • DATASET_DIR = 'path//to//your//dataset'