Neela Harish and S. Poonguzhali. Alphabet Recognition System using Sign Language Gestures. Dynamic Systems and Applications 30 (2021) No.6, 981 – 993 

https://doi.org/10.46719/dsa20213065

ABSTRACT. 
The sign language which is known as visual-gestural language is one of the most important tools for speech-impaired and hearing-impaired people.  The main discrepancy between a sign and a gesture is that, gestures are made at the moment of speaking, but signs are used as an alternate of speaking. There are many recognition systems available for detecting sign languages, despite the fact that their communication difficulties have not yet been solved. It is considered that some devices have a high cost, heavyweight and low recognition rate. So, the aim of this paper is to design and develop the hand recognition framework for speech impaired people, with a high recognition rate and a reduced latency period. The components employed were – flex sensors of 4.5 inches to capture the extent of  bending and the 6-axis accelerometer to detect the angle deflections. By using the controller unit and the wireless transmission unit, the recognized data is sent to the voice processing section using the ZigBee module. Finally, through this system, the speech impaired would be able to spell their name using the gesture of 26 alphabets of the Indian sign language. The efficiency of the proposed setup has been achieved to 95.7% with 10 participants. This will help the user to spell any word continuously by concatenating the letters. As a result, the barrier gap between the normal people and the speech impaired would be diminished to some extent by using this system.

KEYWORDS: Indian Sign Language (ISL), 6 Axis Accelerometer, Flex Sensor, ZigBee Module, Artificial Neural Network (ANN).