Ghanaian Developer Pioneers AI-Powered Assistive Technology Tailored for African Users

Emmanuel Quartey, a Ghanaian IT student, has developed Steward Vision Assistant, an AI-powered platform designed to aid blind and visually impaired individuals. His inspiration stemmed from a personal experience with temporary hearing loss and observing a visually impaired friend's challenges, leading him to adapt computer vision technology for assistive purposes.

Steward Vision Assistant offers features like currency recognition, object identification, text reading, and environmental navigation. It leverages advancements in AI, such as optical character recognition and computer vision, building on existing models to focus on local adaptation rather than developing capabilities from scratch. This approach addresses the significant need for assistive technology in Africa, where over 200 million people require such products, a number projected to double by 2050.

Crucially, Quartey recognized that many imported assistive technologies are not optimized for African contexts. For instance, existing smart glasses often lack support for local currencies like the Ghanaian cedi, and speech recognition models struggle with African accents and languages. Steward aims to overcome these limitations by training its models on Ghanaian accents and exploring support for regional languages like Hausa, Swahili, Twi, and Ga.

Disability inclusion consultant Olamide Daniel highlights the importance of this local focus, noting that imported assistive devices are often expensive and ill-suited for the realities on the continent. Steward's potential to provide an affordable, locally relevant solution could significantly enhance independence for visually impaired Africans, offering capabilities comparable to high-cost international products but with critical regional adaptations.

The platform currently includes AI-powered smart glasses that interpret visual information, providing spoken descriptions of text, currency, objects, and environments. This text-reading feature was particularly inspired by the inaccessibility of updated educational materials for blind students in Ghanaian schools, aiming to bridge the gap left by outdated Braille libraries.