evidoria

← Back to browse

Good practice Imported

Zerobionic — AI-Powered Sign-Language Translation Robots for Deaf Students in Rural Kenyan Classrooms

Kenya · Nairobi · See the Kenya profile · See the Nairobi profile

Evidence: Descriptive / self-reported Top 82% 33/100 · Ask Evidence Copilot about this practice

A Nairobi startup's 3D-printed, recycled-plastic humanoid robots translate speech to sign language in real time for deaf students in low-connectivity Kenyan schools. Self-reported 92% accuracy and rising recognition, but no independent outcome evaluation yet exists.

92% accuracy
Self-reported speech-to-sign translation accuracy (2025-2026)
Zerobionic — AI-Powered Sign-Language Translation Robots for Deaf Students in Rural Kenyan Classrooms

Details

Maturity
Pilot
Promoter
Zerobionic, in partnership with Young Scientists Kenya (YSK)
Period
Founded 2023; classroom piloting since 2025, ongoing into 2026
Keywords
AI robotics, sign language translation, deaf and hard-of-hearing, STEM education, assistive technology

Context

Zerobionic is a Nairobi-based assistive-robotics startup founded by Maxwell Opondo that builds semi-autonomous humanoid robots, 3D-printed from recycled plastic waste to cut costs, using machine vision and machine learning to translate spoken lessons into sign language in real time for deaf students.

Objectives

The robots aim to close the roughly two-hour-per-lesson time penalty deaf students in mainstream science classrooms often face while waiting for human interpretation of equations and concepts, while letting teachers instruct remotely over low-bandwidth 2G connections in schools without reliable internet.

Activities

In partnership with Young Scientists Kenya (YSK), Zerobionic has piloted its robots in schools across Garissa, Machakos and Kirinyaga counties; the robots relay deaf students' signed responses back to teachers as audio or text.

Results

The company reports 92% translation accuracy and early pilot results showing a measurable rise in deaf students' interest in STEM subjects; these figures come from the company itself, not from a third-party or peer-reviewed evaluation. Zerobionic has also drawn recognition, including the Young Tech Innovator Award at the 2025 Africa Tech Summit and selection as Kenya's sole representative in Qualcomm's 2026 Make in Africa Mentorship Programme.

Conclusions

Zerobionic is an early-stage, promising but not yet independently verified practice: it is included as an honest, evidence-labelled example of a low-cost, low-connectivity accessibility approach worth tracking, rather than a proven model to replicate outright, and the company is still seeking investment for mass production.

Implementation

Indicative cost
Low (< €50k) — Recycled-plastic 3D printing is designed to lower unit cost relative to conventional robotics, though no unit price has been published and the company is still seeking investment for mass production.
Time to results
Medium (1–3 years) — Founded 2023; classroom piloting since 2025, ongoing into 2026.
Staffing & skills
Zerobionic founder and engineering team (Maxwell Opondo), Young Scientists Kenya (YSK) as classroom-access partner

Conditions for success

  • 3D-printing from recycled plastic waste lowers unit production cost
  • Low-bandwidth 2G operation allows use in schools without reliable internet
  • Partnership with an established NGO (YSK) provided access to pilot classrooms

Common failure modes

  • 92% accuracy figure and reported STEM-interest gains are self-reported with no independent verification or disclosed methodology
  • Still seeking investment for mass production, limiting scale beyond three counties
  • No documented teacher training programme for structured pedagogical support

Where it fits

Governance type
startup-led with NGO partnership
Scale
pilot across three rural counties
Income level
lower-middle-income (Kenya), targeting low-connectivity rural areas

Commonly funded by

Own resources / municipal budget

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

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.

Attachments

Similar practices you may find useful