Flex Sensor Enabled Smart Sign Language Gloves

Abstract

Communication barriers continue to be a major problem between the general public and people with speech and hearing impairments. A wearable Smart Sign Language Glove was created to solve this problem by translating preset hand gestures into speech and text outputs in real time. The system incorporates an MPU6050 motion sensor to record hand orientation and movement dynamics, as well as three flex sensors to detect finger bending. Sensor data collection, preprocessing, and wireless transmission to a Flask-based backend via Wi-Fi are all managed by an ESP32 Dev Module. Python-based logic is used to implement gesture classification, and a Streamlit web interface with integrated text-to-speech functionality is used to display the recognized output. Near real-time communication is made possible by experimental results showing an average recognition accuracy of about 90% and a response time of almost one second. The system offers an assistive communication solution that is scalable, portable, and reasonably priced.