Design and Development of a Voice-to-Ecuadorian Sign Language Translator to Improve Inclusion of the Deaf in Ecuador

This project addresses the limited communication options for deaf individuals in Ecuador due to the general public's low proficiency in Ecuadorian Sign Language (LSE). To promote inclusion and develop assistive technology, the proposal is to create a software application that translates voice or short texts into LSE via a Virtual Avatar. This avatar will be designed in Blender and animated using Unity. The core of the research focuses on interpreting the input text and its subsequent conversion into signs, necessitating the application and evaluation of Natural Language Processing (NLP) tools to identify the correct signs. Furthermore, two algorithms will be utilized to evaluate semantics and syntax, ensuring accurate translation responses. The system's performance will be assessed using the BLEU metric. This technological development aims to bridge the existing communication gap, which is compounded by a shortage of certified interpreters, by providing a real-time translation tool that serves both communication and active LSE learning purposes.<br/><br/><b>Goal</b>: <br/>The main objective of the project is to design and develop a voice-to-Ecuadorian sign language (LSE) translator to improve the social inclusion of deaf people in Ecuador.<br/><br/><b>Research lines</b>: <br/>Human-computer interaction