AI-POWERED SPEECH-TO-SIGN LANGUAGE TRANSLATION SYSTEM (A CASE STUDY OF NIGERIA)
Abstract
This project presents the design and implementation of an AI-Powered Speech-to-Sign Language Translation System, developed to bridge the communication gap between hearing individuals and persons with hearing impairment in Nigeria. The study was motivated by the limited availability of assistive communication technologies tailored to Nigerian Sign Language (NSL). The system was implemented using Python 3.10 in the Visual Studio Code (VS Code) environment, integrating Tkinter for the graphical user interface, Pillow (PIL) for animated GIF handling, and Pygame for audio playback. The program processes pre-recorded audio phrases, maps them to corresponding NSL animations, and displays the translations sequentially within an interactive graphical interface. Testing and evaluation showed smooth synchronization between speech playback and sign display, demonstrating the effectiveness of the system in translating basic conversational phrases such as greetings and common questions. The prototype achieved high user satisfaction based on its simplicity, responsiveness, and potential educational use. Despite limitations related to dataset size, accent variation, and vocabulary coverage, the system provides a foundation for developing localized AI-driven assistive tools for inclusive communication. Future improvements could include real-time speech recognition, extended NSL datasets, and deployment as a mobile or web-based platform. Overall, the research demonstrates the practical application of artificial intelligence in promoting accessibility, inclusion, and equal participation for the deaf community in Nigeria.
Full-Text Access Notice
In accordance with the NERD Policy on promoting peer-reviewed publication, public access to the full text of a project, thesis or dissertation is restricted for three years, allowing the author and supervisors sufficient time to pursue peer-reviewed publication.
During this period, researchers with legitimate academic or research purposes may request authorisation directly from the author to enable NERD to release the indexed full texts of the work using the form below.