Voice Based Answer Evaluation System for Physically Disabled using Natural Language Processing

Abstract

The proposed work involves the selection of a subject and evaluation of student responses via a voice-based answer evaluation system that utilizes Natural Language Processing (NLP). This system aims to assist physically disabled individuals, who find it challenging to write their answers by hand. Traditional evaluation methods may become time-consuming, biased, and inconsistent in grading. The approach processes spoken answers, converts the voice signal to text, and finds the relevance of these text answers, according to certain criteria based on a predefined marking scheme. Using  NLP techniques ensures maximum grading accuracy with minimum human interaction. The developed system seems to hold promise in accurate evaluation of responses, reduced bias, and greater accessibility for disabled persons.