Smart Wearable Speaking Aid for Aphonic Personnel

Gesturing is an instinctive form of non-verbal or non-vocal communication in which perceptible bodily actions are used to communicate definite messages in conjunction with or without speech. But, aphonic personnel rely solely on this way of communication. This inhibits them from interacting or communicating freely with other people. This paper deals with a Smart Wearable Speaking aid which translates Indian Sign Language into text and voice in real time based on SVM (Support Vector Machine) Model using Raspberry Pi, Flex Sensors, and Accelerometer. Additionally, LCD is used to display the text corresponding to gestures and equivalent output in the form of voice signal is given to the Bluetooth Speaker. The prototype was designed to identify 5 gestures. During the testing of this device, it was observed that the accuracy of the prototype is 90%.

[1]  N. Sriram,et al.  A hand gesture recognition based communication system for silent speakers , 2013, 2013 International Conference on Human Computer Interactions (ICHCI).

[2]  Carlos Emmanuel A. Quiapo,et al.  Development of a sign language translator using simplified tilt, flex and contact sensor modules , 2016, 2016 IEEE Region 10 Conference (TENCON).

[3]  Boon Giin Lee,et al.  Smart Wearable Hand Device for Sign Language Interpretation System With Sensors Fusion , 2018, IEEE Sensors Journal.

[4]  Nello Cristianini,et al.  An introduction to Support Vector Machines , 2000 .

[5]  Sonali N. Pawar,et al.  Hand Glove to Translate Sign Language , 2016 .

[6]  Juhi Ekbote,et al.  Indian sign language recognition using ANN and SVM classifiers , 2017, 2017 International Conference on Innovations in Information, Embedded and Communication Systems (ICIIECS).