This AI technology can understand words that are not even spoken out loud
Researchers at Pohang University of Science and Technology (POSTECH) have developed a Breakthrough Wearable Technology Which can convert silent speech into audible sound by reading the subtle movements of the neck muscles. The study, led by Professor Sung-min Park and Dr. Sunguk Hong, was published in Cyborg and Bionic Systems, marking an important step forward in human-machine communication.
From muscle movements to spoken words
The innovation is built on a simple but powerful idea: Speech is not just about sound. When a person speaks – or even attempts to speak quietly – small movements occur in the muscles and skin around the neck. These activities create a kind of “invisible map” of the intended speech.
To capture this, the researchers created a wearable device called a multiaxial strain mapping sensor. The system combines a miniature camera with flexible silicone embedded with reference markers, allowing it to detect even the smallest deformities of the skin. Designed for daily use, the sensor can be worn comfortably around the neck and automatically recalibrates when reinstalled.
The collected data is then processed using artificial intelligence, which interprets the stress patterns and reconstructs the desired words or sentences. By combining this with voice synthesis trained on the user’s vocal profile, the system can generate speech that closely resembles the person’s natural voice – even when no sound is produced.
A practical leap over existing systems
Traditional voice restoration methods rely on technologies such as electromyography (EMG) or electroencephalography (EEG), which often require bulky equipment and may be inconvenient for long-term use.
The POSTECH team’s approach eliminates these barriers by offering a lightweight, wearable alternative. In testing, the system demonstrated high accuracy in reconstructing speech, even in noisy environments such as industrial settings where traditional microphones struggle.
Real-world impact and future prospects
The implications of this technology are far-reaching. This could provide a new communication route for patients who have lost their voice due to vocal cord damage or laryngeal surgery, enabling them to “speak” again using their own voice profile.
Beyond health care, this system could enable silent communication in environments where speaking out loud is impractical – such as libraries, meetings, or high-noise workplaces. It also opens the door to more naturalness human-ai interface, where intent can be translated into speech without physical intonation.
looking ahead
The researchers aim to refine the technology for widespread deployment in the real world, improving accuracy and expanding language capabilities. Future iterations may integrate more seamlessly with consumer devices, potentially changing the way people communicate in both personal and business settings.
As AI merges wearable technologySuch innovations signal a shift toward more spontaneous, unobtrusive forms of conversation – where even unspoken words can eventually be heard.
