Engineers are focusing on a device measuring one square inch that could help people regain their ability to speak without vocal cords. This could be particularly beneficial for patients with vocal cord pathologies or those recovering from laryngeal cancer surgery, as engineers at the University of California are developing this miniature patch. Last Wednesday, researchers published a paper in the journal "Nature Communications" outlining their invention and the results of testing the device on eight adults. They describe the wearable tool as a self-powered sensing and actuation system based on soft magnetic elasticity, designed to assist in vocalization. The device works by capturing muscle movements made by the patient when speaking (typically using regular speech or mimicking mouth movements). The sensing part of the device recognizes these movements without the need for vocal cord vibrations. The electrical signals are then input into a pre-trained machine learning model, which converts throat movements into sound signals. While there are other solutions available to help individuals without vocal cords produce speech, most of these involve handheld electronic larynges or tracheoesophageal puncture procedures. Both of these solutions can be considered inconvenient and uncomfortable. How does it work? The system is installed on the participant's throat, collecting data on throat muscle movements, and then using machine learning algorithms to associate the generated signals with specific words. The research team had patients say five sentences aloud and silently, including phrases like "I don't trust you" and "I love you." Each person had to repeat these five sentences 100 times for data collection purposes. The overall prediction accuracy of the model was 94.68%, and it was found to be applicable to different populations. Looking ahead, the research team plans to expand the vocabulary of the device through machine learning and test their findings on individuals with speech impairments.