Researchers in South Korea have developed a new sign language translation system based on users wearing seven rings equipped with sensors. According to a new study published in the journal Science Advances, the technology can reliably recognize and translate both American and International sign language words with roughly 88% accuracy.
For many deaf people, sign language is the primary means of communication. Far from being a set of gestures, it is a complex language with its own syntax, grammar, and structure. That makes it difficult for people who don’t sign to communicate with those who do. This challenge is even more complex, given that, according to the World Federation of the Deaf, more than 300 different sign languages are used globally.
Among the current ways to tackle this language barrier are bulky gloves or wired sensor arrays, but these are often uncomfortable, restrict natural hand movements, and require calibration for each user.
A ring system
To solve these problems, scientists developed WRSLT (wirelessly connected, ring-type sign language translator), consisting of seven small rings placed on selected fingers. As the acronym suggests, there are no wires between the rings, allowing full freedom of movement and natural signing.
Each ring contains a tiny accelerometer, which is the same type of device that tells your smartphone to rotate its screen when you turn it. When a person signs while wearing the rings, sensors measure finger orientation relative to gravity and capture the direction of hand motion.
These signals are then sent to a smartphone or a computer via Bluetooth multilink technology. Finally, AI interprets these movements and gestures into written text. “The WRSLT enables real-time, sentence-level translation using a fully wireless and modular architecture that generalizes across users and sign languages,” wrote the team in their paper.
Currently, WRSLT can recognize 100 words in American Sign Language (ASL) and 100 words in International Sign Language (ISL).
Proving the concept
The system was trained on a set of users and then tested on others who were not included in the training. It correctly recognized ISL words with 88.5% accuracy and ASL words with 88.3% accuracy. “Despite interuser variability and potential differences in sensor alignment, the system achieved high recognition accuracies… validating the generalizability and robustness of the proposed platform,” according to the researchers.
Next up for the research team is to further miniaturize the rings and expand the AI training data to include more words and different sign languages.
Written for you by our author Paul Arnold, edited by Gaby Clark, and fact-checked and reviewed by Robert Egan—this article is the result of careful human work. We rely on readers like you to keep independent science journalism alive.
If this reporting matters to you, please consider a donation (especially monthly). You’ll get an ad-free account as a thank-you.
Publication details
Jaejin Park et al, An AI-driven, wearable, conformal ring system for real-time and user-independent sign language interpretation, Science Advances (2026). DOI: 10.1126/sciadv.aec8995
Key concepts
The content is provided for information purposes only.