Prototipo de ojo de lectura para convertir texto impreso y escrito a voz sintetizada con uso de un ordenador de IA.
Fecha
2024-02-29
Autores
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad de Guayaquil. Facultad de Ingeniería Industrial. Carrera de Ingeniería en Teleinformática
Resumen
El propósito de la presente investigación es la realización de un prototipo de ojo de lectura, capaz de traducir los textos impresos y escrito a voz sintetizada en tiempo real. Para ello se hará uso de la visión artificial, reconocimiento de caracteres y Python, la necesidad de este prototipo radica en obtener una alternativa eficaz en cuanto al aprovechamiento de texto para personas con discapacidad visual, para su desarrollo se indago en las necesidades del usuario, se escogió un motor OCR que permitirá la conversión de texto y un micro procesador acorde a las necesidad a cubrir, luego el sistema traductor de textos fue portado al prototipo Jetson Nano, donde se hará ejecución en un entorno virtual que contenga previamente bibliotecas específicas para su funcionamiento. Finalmente, se presentarán pruebas realizadas en diferentes escenarios para verificar la funcionalidad del prototipo, observándose que la eficiencia de inferencia en pantalla alcanzó el 80%.
The purpose of this research is the creation of a prototype of a reading eye, capable of translating printed and written texts into synthesized speech in real time. For this, artificial vision, character recognition and Python will be used. The need for this prototype lies in obtaining an effective alternative in terms of the use of text for people with visual disabilities. For its development, the user's needs were investigated. An OCR engine was chosen that will allow text conversion and a micro processor according to user needs, then the text translation system was ported to the Jetson Nano prototype, where it will be executed in a virtual environment that previously contains specific libraries for its functioning. Finally, tests carried out in different scenarios will be presented to verify the functionality of the prototype, observing that the inference efficiency on the screen reached 80%.
The purpose of this research is the creation of a prototype of a reading eye, capable of translating printed and written texts into synthesized speech in real time. For this, artificial vision, character recognition and Python will be used. The need for this prototype lies in obtaining an effective alternative in terms of the use of text for people with visual disabilities. For its development, the user's needs were investigated. An OCR engine was chosen that will allow text conversion and a micro processor according to user needs, then the text translation system was ported to the Jetson Nano prototype, where it will be executed in a virtual environment that previously contains specific libraries for its functioning. Finally, tests carried out in different scenarios will be presented to verify the functionality of the prototype, observing that the inference efficiency on the screen reached 80%.
Descripción
Este documento contiene archivo en PDF
Palabras clave
DISCAPACIDAD VISUAL, VOZ SINTETIZADA, JETSON NANO, VISIÓN ARTIFICIAL