UKZN graduate develops AI system that translates SA Sign Language into speech - IOL
The vision? Akhil Hansrajh hopes to help address communication barriers faced by people who are hard of hearing.
Hansrajh developed the SASL-to-Speech Translation System as his final-year design project while completing his BSc Engineering in Computer Engineering.
The computer vision-based system uses a custom-built artificial intelligence model alongside an established machine learning framework to track body landmarks and identify signs in real time.
It was trained using a custom dataset relevant to SASL and can recognise individual letters and words before converting the interpreted signs into speech through a text-to-speech model.
According to Hansrajh’s supervisor, Dr Sulaiman Patel, the project was among the first assistive technology projects in UKZN’s Electrical, Electronic and Computer Engineering (EECE) discipline to focus specifically on the South African dialect of sign language.
SASL was recently recognised as South Africa’s 12th official language, while many comparable sign-language translation systems internationally focus on American Sign Language.
For Hansrajh, the project had a deeply personal motivation. Both his parents were born hard of hearing and have experienced communication barriers in everyday situations that is often taken for granted.
“My parents encounter communication barriers in everyday settings such as banks, government departments, and municipal offices,” said Hansrajh. “My sister and I have often had to accompany them to speak on their behalf.”
Hansrajh said a fully developed version of the technology could help people communicate more independently.
“A production-ready system would empower these individuals to communicate more easily and effectively, without requiring the other party to have learned SASL,” he said.
While a fantastic development, Hansrajh's road to graduation was not without challenges.
Patel said Hansrajh experienced mental health challenges during the second semester of 2025 and required an extension for his first project submission.
With Patel’s support, including referrals to student counselling and regular wellness check-ins, Hansrajh was able to regain momentum and continue with the project.
His efforts ultimately earned him a highly commended award from the EECE for his final-year project.
Hansrajh said he particularly enjoyed the design modules during his undergraduate studies because they allowed him to apply theoretical knowledge to substantial projects.
He also identified Data Communications and Internet Engineering, taught by Professor Tahmid Quazi, as his favourite modules.
With one outstanding module to complete in 2026 and facing financial difficulties, Hansrajh was recommended by Quazi to a local company for an interview. He secured a job and worked full-time while completing his studies.
Now that he has completed his BSc Engineering degree, Hansrajh plans to continue developing his sign-language translation project.
His focus will be on improving the system’s ability to recognise words and produce more cohesive and fluent speech.
In the longer term, he hopes to pursue a career in software, networking, or artificial intelligence and machine learning.
Hansrajh credited his parents, uncle, sister, Patel, friend Aamil Maharaj, as well as his roommate, girlfriend and university colleagues, for supporting him throughout his journey.
Outside his studies, he enjoys gaming, following developments in technology and working on programming side projects. He also thanks UKZN for the lessons he learned and the opportunity to experience university life.
