High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

LEAP for DEAF : A Sign Language to speech translator

Author(s):

Kadam Bhushan Ravindra , NDMVP's KBTCOE, Nashik; Kadam Bhushan Ravindra, NDMVP's KBTCOE, Nashik; Supekar Komal Kalidas, NDMVP's KBTCOE, Nashik; Chikhale Harshad Kailas, NDMVP's KBTCOE, Nashik; Salve Priyanka Balasaheb, NDMVP's KBTCOE, Nashik

Keywords:

Leap Motion Sensor, Gesture Recognition, Indian Sign Language, Human Computer Interaction, Machine Learning

Abstract

Sign language is vital for facilitating communication between hearing impaired and the rest of society. Researchers in Sign language recognition had tailored completely different sensors to capture hand signs. Gloves, digital cameras, depth cameras and Kinect were used instead in most systems. Owing to signs closeness ,input accuracy is a terribly essential constraint to achieve a high recognition accuracy. Our aim is to design a Sign Language to Speech Translation system for ISL(i.e. Indian Sign Language) based on a brand new digital 3D motion detector referred to as Leap Motion, that consists of two inbuilt cameras and three infrared sensors that capture 3-D dynamic hand gestures. The palm-sized Leap Motion sensor provides far more portable and economical solution than Cyber glove or Microsoft Kinect employed in existing studies . This sensor tackles the major issues in vision-based systems like skin color, lighting etc... The planned system will make use of DTW algorithm as a classifier for converting hand gestures into an appropriate text as well as an audible speech.

Other Details

Paper ID: IJSRDV4I20929
Published in: Volume : 4, Issue : 2
Publication Date: 01/05/2016
Page(s): 1054-1059

Article Preview

Download Article