Researchers at the University of Cambridge have developed a new wearable device, named Revoice, that could help stroke patients regain their ability to speak without the need for invasive brain implants. The device, worn as a soft choker around the neck, uses sensors and artificial intelligence to decode speech signals and emotional cues, translating them into words and complete sentences.
Revoice works by picking up the patient's heart rate and tiny vibrations from throat muscles. These signals are then rapidly translated into speech. The AI can also interpret emotional state and contextual information, such as the time of day or weather, to predict the intended sentence. In a small trial involving five patients with dysarthria—a speech difficulty often following a stroke—the device achieved high accuracy, with only 4.2% of words and 2.9% of sentences incorrect.
Professor Luigi Occhipinti, who led the research, said the device helps both patients and therapists establish more effective therapy for recovering speech. He noted that while therapy helps patients regain words through repetition, they often struggle when asked to speak freely, leading to frustration. Revoice aims to give patients their independence back, as communication is fundamental to dignity and recovery.
The device is designed to be comfortable and usable outside hospital settings, allowing patients to gradually recover language function at home. Users can also control the device by nodding twice to expand words into sentences. The study, published in Nature Communications, reported a 55% increase in satisfaction among participants.
Juliet Bouverie, chief executive of the Stroke Association, welcomed the early-stage research but cautioned that the findings are preliminary, as the device has only been tested with a small number of people. She emphasised that specialist stroke rehabilitation remains essential, but innovations like Revoice could complement existing support and help more stroke survivors regain independence.
