SpellRing

AI powered ring converts American Sign Language into text for smartphones and computers.
1
2
3
4
5
images courtesy of the researchers from Cornell University, Carnegie Mellon University, Stony Brook University
images courtesy of the researchers from Cornell University, Carnegie Mellon University, Stony Brook University
images courtesy of the researchers from Cornell University, Carnegie Mellon University, Stony Brook University
images courtesy of the researchers from Cornell University, Carnegie Mellon University, Stony Brook University
images courtesy of the researchers from Cornell University, Carnegie Mellon University, Stony Brook University
images courtesy of the researchers from Cornell University, Carnegie Mellon University, Stony Brook University
images courtesy of the researchers from Cornell University, Carnegie Mellon University, Stony Brook University
images courtesy of the researchers from Cornell University, Carnegie Mellon University, Stony Brook University
images courtesy of the researchers from Cornell University, Carnegie Mellon University, Stony Brook University
images courtesy of the researchers from Cornell University, Carnegie Mellon University, Stony Brook University
Previous Pause Next

A team of researchers from Cornell University, in collaboration with Carnegie Mellon University and Stony Brook University, have developed SpellRing, an innovative AI-powered wearable ring that translates American Sign Language (ASL) fingerspelling into text for smartphones and computers.

Believed to be the first of its kind, SpellRing utilizes sonar technology to track hand and finger gestures in real-time, seamlessly converting them into digital text communication. The ring is equipped with a microphone and speaker to emit and receive inaudible sound waves to analyze motion with precision. Additionally, a mini-gyroscope tracks overall hand gestures while a deep-learning algorithm processes this amassed data to recognize fingerspelled words. 

Recent trials involving 20 ASL signers demonstrated accuracy rates between 82% and 92%, depending on word complexity. This impressive performance highlights SpellRing’s potential as a lightweight, ergonomic, and unobtrusive solution for assistive sign language translation.