Review of Intelligent Methods for Sign Language-to-Speech Translation

In today's digital age, technology has significantly transformed communication. However, for the deaf community, the communication barrier remains a persistent challenge. This article addresses the critical need to develop intelligent systems that translate sign language into speech, highlighting its potential to improve the social inclusion of people with hearing impairment. The study conducts a systematic review of advanced methods, such as the use of convolutional neural networks (CNNs) and recurrent neural networks (RNNs), to identify promising techniques and persistent challenges in this field. The research is based on the analysis of 32 articles selected using strict inclusion and exclusion criteria, following the PRISMA statement. The results reveal that these systems can improve the autonomy of deafblind people, facilitating their access to essential services and promoting their social and occupational inclusion. Furthermore, it highlights the need for solutions adapted to specific local contexts due to the geographical variability in the prevalence of hearing impairment. This work not only contributes to scientific knowledge, but also lays the foundation for the development of more effective and accessible technologies, potentially transforming areas such as education, healthcare and employment for the deaf community.