This project aims to provide a real-time sign language translation system using deep learning. The system recognizes hand gestures captured by a webcam and translates them into text, displaying the corresponding meaning in the chosen language.
- Language Selection: Users can choose their desired language from a list of options, including Hindi, Marathi, Gujarati, Kannada, Tamil, Telugu, Bengali, and Malayalam.
- Translation: The system translates recognized sign language gestures into the selected language.
- Font Customization: The translated text is displayed using custom fonts suitable for each language.
- Sign Language Models: The project includes three different sign language models for American Sign Language (ASL), Indian Sign Language (ISL), and general hand gestures.
- Main script for the graphical user interface (GUI) and real-time sign language translation.
- Utilizes the Tkinter library for GUI development.
- Implements a webcam interface for capturing hand gestures.
- Allows users to choose the target language and submit their selection for translation.
- Includes buttons to choose between ASL, ISL, and general gestures for translation.
- Script for training the American Sign Language (ASL) model.
- Uses TensorFlow and Keras for deep learning.
- CNN architecture for recognizing hand gestures.
- Data augmentation and preprocessing are applied using ImageDataGenerator.
- The trained model is saved as
asl_model.h5.
- Script for training the Indian Sign Language (ISL) model.
- Similar structure to ASL training script.
- The trained model is saved as
isl_model.h5.
- Script for training a model to recognize general hand gestures.
- The model can identify gestures like "Bye," "Call Me," "Dislike," "Good Job," "Good Luck," "Peace," "Praying," and "Rock On."
- The trained model is saved as
model.h5.
- Script for capturing and creating a dataset for training.
- Uses webcam input to capture hand gestures.
- Segments the hand region and saves images for training.
- Execute
Final_deployement.pyto launch the application. - Choose the target language and click "Submit" to start real-time sign language translation.
- Use the buttons to select the sign language model for translation.
- TensorFlow
- Keras
- OpenCV
- Numpy
- Googletrans
- Tkinter
- PIL
For any questions or feedback, feel free to reach out:
- Siddhartha Chakrabarty - GitHub | LinkedIn
- Sneha Jain - GitHub | LinkedIn
- Project Repository
This project was developed as part of a deep learning and computer vision exploration.
Feel free to contribute and improve this project!