NATURAL LANGUAGE PROCESSING AND TEXT-TO-SPEECH TECHNOLOGY

Ademi, Valbon and Ademi, Lindita (2023) NATURAL LANGUAGE PROCESSING AND TEXT-TO-SPEECH TECHNOLOGY. Journal of Natural Sciences and Mathematics of UT, 8 (15-16). pp. 299-306. ISSN 2545-4072

Text JNSM 2023-299-306.pdf Download (637kB)

Abstract

Text-to-speech (TTS) technology is the process by which the computer is made to speak. It uses natural language processing concepts. Despite the advancement of technology that allows information to be stored electronically, textual information remains the most common way of exchanging information. Using text documents is problematic for visually impaired people in many scenarios, such as reading text on the move and accessing text under less-than-ideal conditions. The goal is to allow blind users to touch the printed text and receive the real-time transmission of the words. The development of such systems requires the use of such systems, requires the use of two technologies that are central to these systems, namely optical character recognition (OCR) to extract text information (Text Information Extraction) and text-to-voice (TTS) to convert this text in question. Text information extraction is the first and most important function of any assistive reading system and is an integral part of OCR because this process determines the intelligibility of the extracted word. Recent developments in computer vision, digital cameras, and computers make it possible to develop cameras and products that combine computer vision technology with other existing useful products such as optical character recognition systems used to recognize words. She can recognize characters, words, and sentences without any mistakes. OCR has a high recognition rate which is the electronic conversion of photographed images of typed or typed text into computer-readable text. Developments in computer technology make it possible to help these individuals by developing camera-based products. People with poor vision need portable assistance to read this printed text. The need to develop a voice-assisted text-to-speech system using the optical character recognition method with different sets of input and speech output is simulated.

Item Type:	Article
Uncontrolled Keywords:	language, processing, synthesis, TTS, voice, visually impaired.
Subjects:	T Technology > T Technology (General)
Divisions:	Faculty of Engineering, Science and Mathematics > School of Electronics and Computer Science
Depositing User:	Unnamed user with email zshi@unite.edu.mk
Date Deposited:	04 Nov 2023 23:17
Last Modified:	04 Nov 2023 23:19
URI:	http://eprints.unite.edu.mk/id/eprint/1528

Actions (login required)

View Item