A wearable device could help people regain the ability to communicate naturally and fluently following a stroke, without the need for invasive brain implants.
The device, named Revoice, uses a combination of ultra-sensitive sensors and artificial intelligence to decode speech signals and emotional cues to allow people with dysarthria, a post-stroke speech impairment, to communicate.
Worn as a soft and flexible choker, the device captures the wearer’s heart rate and tiny vibrations from throat muscles, and uses those signals to reconstruct intended words and sentences in real time.
The signals from the device are processed by two AI agents: one reconstructs words from fragments of silently mouthed speech, while the other interprets emotional state and contextual information, such as the time of day or weather conditions, to expand short phrases into complete, expressive sentences.
The development was led by researchers at the University of Cambridge.
Professor Luigi Occhipinti from Cambridge’s Department of Engineering, said: “When people have dysarthria following a stroke, it can be extremely frustrating for them, because they know exactly what they want to say, but physically struggle to say it, because the signals between their brain and their throat have been scrambled by the stroke.”
The professor explained that while most stroke patients are able to perform repetitive word drills when working with a speech therapist, they often struggle with open-ended questions and everyday conversation.
“And as many patients do recover most or all of their speech eventually, there is not a need for invasive brain implants, but there is a strong need for speech solutions that are more intuitive and portable,” he said.
Working with colleagues in China, the researchers carried out a small trial with five stroke patients with dysarthria, as well as ten healthy controls. In the study, participants wore the device and mouthed short phrases. By nodding twice, they could choose to expand those phrases into sentences using the embedded lightweight large language model (LLM).
In one example, “We go hospital” became “Even though it’s getting a bit late, I’m still feeling uncomfortable. Can we go to the hospital now?” The sensors in the Revoice device inferred that the wearer was feeling frustrated due to their elevated heart rate, and that it was getting late at night. The LLM was able to use this data to expand three mouthed words into a full sentence.
Although extensive clinical trials will be required before the device can be made widely available, the researchers hope that future versions of the device will include multilingual capabilities, a broader range of emotional states and fully self-contained operation for everyday use.
“This is about giving people their independence back,” said Occhipinti. “Communication is fundamental to dignity and recovery.”
The researchers say the technology could aid stroke rehabilitation and also support people with conditions such as Parkinson’s and motor neurone disease.
They are currently planning a clinical study in Cambridge for native English-speaking dysarthria patients to assess the viability of the system, which they are hoping to launch this year.