Design and Evaluation of SmarText PDF: An Automated Proofreading System with Text-to-Speech Functionality

Read full text
Authors: Kim M. Nebreja, Jayle Kristel P. Posadas, Cyrus Lander V. Alim
Cited by: 0

Abstract

The use of PDF documents across academic and professional environments hasunderscored the need for efficient, accessible, and intelligent proofreading tools.Traditional methods of proofreading are often time-consuming, possible to human error,and lack support for users with specific accessibility needs such as low vision or auditorylearning preferences. This study presents SmarText PDF: Automated Proofreading withText-to-Speech Functionality, a web-based platform designed to enhance the quality andaccessibility of PDF-based communication. The system leverages Natural LanguageProcessing (NLP) to automatically detect and correct grammatical errors andmisspellings, while providing context-aware suggestions tailored to the document’s toneand intent. It also integrates a text-to-speech (TTS) feature that allows users to listen totheir proofread content, improving comprehension and inclusivity. Additionally,SmarText PDF features a document comparison tool for tracking revisions and offers theability to download corrected files while preserving the original layout. Using theModified Waterfall Methodology, the system was planned, designed, and developed toprovide users with their need for automated proofreading process and accessibility.Evaluation results show that SmarText PDF outperforms traditional proofreading toolsacross all FURPS criteria, with students and experts evaluating the system. The findingsconfirm improvements in accuracy, usability, and reliability, indicating that SmarTextPDF effectively addresses common challenges in manual proofreading and provides anefficient, accessible, and user-friendly solution for enhancing document quality.

Fields of Study & Subjects

Writing and Handwriting Education Text Readability and Simplification Usability and User Interface Design

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
9 January 2026
Language
EN
Document type
Thesis