After five years of pioneering work to design a real-time sign language translator alongside CESAR engineers and researchers, Lenovo opened its annual Tech World ‘23 conference in Austin, Texas last week with the first live, public demonstration of the innovation last week.
It served as the opening moment to greet event-goers for the keynote address from Lenovo Chairman and CEO Yuanquing Yang to shine a spotlight for a one-of-a-kind line-up of industry innovators and global CEOs bringing Lenovo’s vision of A.I. for All to life for the crowd there.
The grand public unveiling starts with a Brazilian software developer named Gabriel Lima greeting the jam-packed conference hall silently using Libras, the official Brazilian Sign Language. Following a short intro video, Gabriel comes back on stage to sign again – but this time to demonstrate the technology that uses computer vision and an original A.I. engine designed in collaboration with CESAR over five years of RD&I development.
The language barrier between Gabriel and the audience instantly disappears as a camera captures the precise movements of Gabriel’s hands and the solution translates them into real-time text and voice in English for the audience. Watch the astonishing moments of the conference’s opening here.
The RD&I team built the new A.I. solution to facilitate interactions where a sign language interpreter may be unavailable—from retail spaces to hospitals—to increase autonomy and create connection. Dozens of deaf and hard of hearing individuals using Libras contributed thousands of hours of anonymized video data to build the training set and advance the A.I., which will eventually be trained on understanding other sign languages from around the world.
Achieving real-time video capture and translation between languages presents a staggering amount of data—not the least of which are the individual gestures for each word and the syntax of each sentence. Just as regional accents within a spoken language like English can be dramatically different, movements and styles can be distinct to individuals within Libras.
To tackle the data challenge, Lenovo collaborated with Brazilian innovation center CESAR, sharing expertise on capturing and cataloging video to lay the foundation for the AI. Lenovo and CESAR have since created a dataset of thousands of Libras videos to train the core algorithm to identify and contextualize individual gestures. Then Lenovo led the way to develop the breakthrough A.I. at the heart of the solution.
The initiative between CESAR and Lenovo is a world-first, with potential for a universal application. Ultimately, the system aims to foster active participation of deaf individuals as communicators rather than just recipients, breaking down existing barriers and promoting a more inclusive society. Read more about this Lenovo | CESAR RD&I collaboration project here via Forbes.
Have a big technical challenge? Contact CESAR to help
CESAR takes immense pride in being part of Lenovo’s A.I. for All, a groundbreaking solution promoting inclusion. This innovation, the culmination of five years of collaborative work, reduces language barriers and empowers individuals like Gabriel to communicate effortlessly with the world. This moment reminds us that with determination and cooperation, we can achieve remarkable feats.