AI-Driven Neckband Enables Silent Communication in Noisy Environments

A team of researchers at Pohang University of Science and Technology has developed an innovative AI-powered neckband that decodes silent speech, offering a breakthrough solution for communication in environments with high noise levels. This new device utilizes advanced sensors and artificial intelligence to interpret throat muscle movements without relying on sound, addressing long-standing challenges in silent speech interface (SSI) technology.

The system incorporates a Computer Vision-Based Optical Strain (CVOS) sensor embedded in a flexible neck choker. This sensor tracks both the magnitude and direction of throat muscle movements during speech with high accuracy, capturing two-dimensional strain maps to provide a comprehensive understanding of complex muscle dynamics. Unlike traditional sensors, which often face limitations such as discomfort and invasiveness, this new technology is designed to be non-invasive and user-friendly.

The AI-driven processing pipeline enables rapid and accurate speech decoding by recognizing both localized muscle movements and broader speech patterns. Remarkably, the system can reconstruct a user’s unique voice using just 10 minutes of recorded speech, making it highly adaptable for various communication needs. Currently focused on recognizing the NATO phonetic alphabet, the device has demonstrated impressive accuracy in laboratory testing, achieving 85% success under controlled conditions and maintaining strong performance even in noisy environments up to 90 decibels.

Beyond industrial and military applications, this technology holds significant potential in healthcare, particularly for individuals with speech impairments. It offers a non-invasive communication method for those who have undergone laryngectomies or other conditions that affect their ability to speak. Future research will aim to expand the system’s vocabulary, enhance resistance to motion-related artifacts, and refine its design for broader usability.

This study, published in *Cyborg and Bionic Systems*, represents a major advancement in SSI technology, paving the way for more effective communication solutions in challenging environments.