High-Tech Patch Provides a Voice to the Voiceless

Engineers at the University of California, Los Angeles (UCLA) have developed a wearable device that – through the power of artificial intelligence (AI) – could help people with vocal cord disorders regain their ability to speak.

The device, consisting of a thin, flexible patch a little over 3 centimetres wide, adheres to the throat, near the vocal cords. Using double-sided biocompatible tape, it translates laryngeal muscle movements into audible speech.

It detects mechanical forces generated by muscle movements, converts them into electrical signals and uses an AI algorithm to predict what the user intends to say. The system then plays a pre-recorded sentence through an embedded speaker.

Jun Chen, an assistant professor of bioengineering at UCLA, leads the Wearable Bioelectronics Research Group, which designed the patch to be a non-invasive alternative to traditional speech aids like electrolarynx devices.

In tests, volunteers silently mouthed five sentences while wearing the device, achieving a 95% accuracy rate in speech prediction. Although the current version is limited to a small set of sentences, researchers plan to expand its vocabulary using advanced AI models. Future iterations may even allow real-time translation of laryngeal movements into full speech.

The patch is lightweight, reusable and designed to assist individuals recovering from surgery, laryngitis or vocal cord paralysis. Unlike existing solutions, it eliminates the need for invasive procedures or cumbersome external devices.

Chen’s team is now focusing on refining the algorithm and testing the device on individuals with speech impairments, aiming to provide a practical solution for those struggling with voice disorders.

Image Credit: Source