Nigerian-UK artificial intelligence company Talksign has introduced two new AI-powered models that can translate American Sign Language (ASL) into text or speech in real time, and vice versa.
The company launched Palm 1.0, which interprets sign language into text or spoken words with 84.2% semantic accuracy, and Echo 1.0, which converts spoken or written language into realistic sign language videos using digital avatars. Both tools work with very little delay, making communication smoother and more natural.
Talksign says the new models are designed to help millions of people with hearing impairments communicate more easily in everyday situations whether in schools, hospitals, workplaces, or public spaces. The technology could be particularly useful in Nigeria and other African countries where access to sign language interpreters is often limited.
The founders described the launch as a step towards greater inclusion and breaking down barriers faced by the deaf and hard-of-hearing community. By combining AI with sign language, Talksign aims to reduce dependence on human interpreters and make communication more accessible and affordable.
Industry experts have welcomed the development, noting that real-time translation tools like these can significantly improve quality of life and opportunities for people living with hearing loss. However, some have called for further testing to ensure high accuracy across different accents, speeds, and cultural contexts.
Talksign, which has offices in Nigeria and the United Kingdom, plans to expand the technology to more sign languages, including Nigerian Sign Language, in the near future.
This innovation adds to the growing number of African-led AI solutions addressing local and global challenges. For many in the deaf community, tools like Palm and Echo could open up new possibilities for education, employment, and social interaction.
The models are now available for testing and integration by developers and organisations.
English 
简体中文 
Čeština 
Deutsch (Sie) 
Français 
