DYNAMIC SIGN LANGUAGE INTERPRETER

DYNAMIC SIGN LANGUAGE INTERPRETER

Anushree C K1, Prof. Sowmya M S2

1 Student, Department of MCA, Bangalore Institute of Technology, Karnataka, India

2 Professor, Department of MCA, Bangalore Institute of Technology, Karnataka, India

---------------------------------------------------------------------***----------------------------------------------------------------------

ABSTRACT

Sign Language Interpreter (SLI) interprets computer vision with machine learning, mainly using K-Nearest Neighbors and Support Vector Machine algorithms in gesture recognition. The SLI detects and analyzes hand gestures along with some other characteristics of sign language communicators. KNN classifies these gestures based on their similarity to known labeled gestures; hence, it can handle multidimensional data. Being a more advanced technique, SVM finds the best possible hyperplane that can separate classes of gestures, hence giving razor-sharp demarcations between different gestures. The performance of the SVM was better in this application than KNN, thus showing its high ability to handle the complexity of sign language gestures, hence improving translation and interpretation.

Keywords: Sign language, KNN, SVM, Gesture-to-Text, Text-to-Speech, Machine Learning.