Echo 🤟 — Every Gesture Has a Voice
An assistive smart-glove system that bridges communication barriers by translating Sign Language gestures into speech using flex sensors, an ESP32, and machine learning. Built to make communication more accessible, with motion-based recognition and neural networks planned for future development.
Team
- E/22/074, Yasas Dewshan, e22074@eng.pdn.ac.lk
- E/22/061, Sayuru Dassanayake, e22061@eng.pdn.ac.lk
- E/22/233, Teshini Matharaarachchi, e22233@eng.pdn.ac.lk
- E/22/253, Tharusha Nethmina, e22253@eng.pdn.ac.lk
Supervisors
- Ms. Yasodha Vimukthi, yasodhav@eng.pdn.ac.lk
Table of Contents
- Introduction
- Solution Architecture
- Hardware & Software Designs
- Testing
- Detailed Budget
- Conclusion
- Links
Introduction
Echo is an assistive smart-glove system designed to translate Sign Language gestures into real-time speech and text. Communication between deaf or hard-of-hearing individuals and non-signers remains a significant challenge. Echo aims to bridge this gap through wearable technology integrating flex sensors, an ESP32 microcontroller, and machine learning models for accurate gesture interpretation.
Solution Architecture
High level diagram + description
Hardware and Software Designs
- Sensors & Processing: Flex sensors mapped to finger bend angles, processed via ESP32.
- Machine Learning & Gesture Recognition: Classification models to identify static and dynamic signs.
- Software Platforms: Web and Mobile Application interfaces for audio playback and visualization.
Testing
Testing done on hardware and software, detailed + summarized results.
Detailed Budget
All items and costs:
| Item | Quantity | Unit Cost | Total |
|---|---|---|---|
| Sample item | - | - | - |
Conclusion
What was achieved, future developments (e.g., motion-based recognition with IMUs and neural networks), and commercialization plans.