Vinsighte's Visis app uses AI-powered OCR and computer vision tuned to African accents to read text aloud for visually impaired students; a UN-documented rollout across 7 Nigerian institutions reports over 90% reading independently.
8000 people
People reached by the Visis deployment (Dec 2020 - Jul 2023)
7 schools
Schools for the visually impaired reached (Dec 2020 - Jul 2023)
4 states
Nigerian states covered
90 %
Students able to independently use the app to read printed materials
Founded in Lagos in 2020 by Oluwatomisin Kolawole and Oyolola Caleb, Vinsighte built Visis, a mobile app using AI-powered OCR and computer vision to read printed text aloud and help visually impaired users identify objects, colours and surroundings, tuned specifically to African accents and multiple local languages, unlike most mainstream screen readers.
Objectives
Open printed educational materials to visually impaired students who previously could only access them with a human reader.
Activities
Between December 2020 and July 2023, Vinsighte deployed Visis across seven schools for the visually impaired in four Nigerian states, reaching more than 8,000 people, with support from the ACT Foundation, Merck, the Rotary Club, the Roddenberry Foundation, GIZ, Orange Corners and MIT Solv[ED], documented through the UN's SDG Partnership Platform.
Results
Feedback collected from students and teachers across the seven institutions found that more than 90% of students could independently use the app to read printed educational materials they previously could only access with a human reader.
Conclusions
The reported outcomes are drawn from partner and program self-reporting rather than an independent third-party evaluation, and public information does not disclose total funding raised - a gap that formal impact evaluation could help close as the tool expands beyond its initial Nigerian deployment sites.
Implementation
Indicative cost
Low (< €50k)
Time to results
Medium (1–3 years)
Staffing & skills
Vinsighte founders (Oluwatomisin Kolawole, Oyolola Caleb), Partner schools for the visually impaired, Funders: ACT Foundation, Merck, Rotary Club, Roddenberry Foundation, GIZ, Orange Corners, MIT Solv[ED]
Conditions for success
AI/OCR tuned to African accents and local languages rather than relying on mainstream (Western-tuned) screen readers
Multi-funder philanthropic/development support enabling free or low-cost deployment
Direct partnership with schools for the visually impaired for real-world testing and feedback
Common failure modes
Reported outcomes rely on partner/program self-reporting rather than independent evaluation
Total funding raised is not publicly disclosed, limiting transparency
No independent accuracy testing of the OCR/computer-vision models has been published
Where it fits
Governance type
social enterprise / NGO-backed startup
Scale
sub-national (7 schools, 4 states)
Income level
lower-middle-income (Nigeria)
Do you run this practice?
Claim it —
verified implementers get a public contact pathway and can propose corrections.
Data sources
Where this practice's information was retrieved from, and when.
STMicroelectronics Foundation, ITU and a blind Milan professor run train-the-trainer digital-literacy classes for blind and visually-impaired adults in Dakar. An …
WeSign gives deaf and hard-of-hearing Hanoi primary pupils real-time, AI-scored feedback on fingerspelling and sign formation via webcam. A 30-participant …
OLE Nepal's 'intelligent OCR' (iOCR) converts scanned Nepali/English documents into screen-reader-compatible e-books, filling a gap in Nepali-script OCR. Built with …