People with voice disorders, including those with pathological conditions of the vocal cords or who are recovering from laryngeal cancer surgeries, may often have difficulty or inability to speak. That may change soon.
A team of UCLA engineers has invented a soft, thin, stretchy device measuring just over 1 inch square that can be placed on the skin outside the throat to help people with dysfunctional vocal cords regain vocal cord function. voice. Its progress is detailed this week in the magazine. Nature Communications.
The new bioelectric system, developed by Jun Chen, assistant professor of bioengineering at the UCLA Samueli School of Engineering, and his colleagues, is capable of detecting movement in the muscles of a person’s larynx and translating those signals into audible speech with the help of machines. learning technology, with almost 95% accuracy.
The breakthrough is the latest in Chen’s efforts to help people with disabilities. Her team previously developed a wearable glove capable of translating American Sign Language into English in real time to help ASL users communicate with those who do not know sign.
The small new patch-like device is made up of two components. One, a self-powered sensor component, detects and converts signals generated by muscle movements into high-fidelity analyzable electrical signals; These electrical signals are then translated into voice signals using a machine learning algorithm. The other, a performance component, converts those speech signals into the desired speech expression.
Each of the two components contains two layers: a layer of biocompatible silicone compound polydimethylsiloxane, or PDMS, with elastic properties, and a magnetic induction layer made of copper induction coils. Sandwiched between the two components is a fifth layer containing PDMS mixed with micromagnets, which generates a magnetic field.
Using a soft magnetoelastic sensing mechanism developed by Chen’s team in 2021, the device is capable of detecting changes in the magnetic field when it is altered as a result of mechanical forces, in this case, the movement of the laryngeal muscles. Serpentine induction coils embedded in the magnetoelastic layers help generate high-fidelity electrical signals for sensing purposes.
The device measures 1.2 inches on each side, weighs about 7 grams, and is just 0.06 inches thick. With the double-sided biocompatible tape, it can be easily adhered to a person’s throat near the location of the vocal cords and can be reused by reapplying the tape as needed.
Voice disorders are prevalent across all ages and demographic groups; Research has shown that almost 30% of people will experience at least one of these disorders in their lifetime. However, with therapeutic approaches, such as surgical interventions and voice therapy, voice recovery can last from three months to a year, and some invasive techniques require a significant period of mandatory postoperative voice rest.
“Existing solutions, such as wearable electrolarynx devices and tracheoesophageal puncture procedures, can be inconvenient, invasive or uncomfortable,” said Chen, who directs the Wearable Bioelectronics Research Group at UCLA and has been named one of the researchers. most cited in the world five years in a row. “This new device presents a portable, non-invasive option capable of helping patients communicate during the pre-treatment period and during the post-treatment recovery period for voice disorders.”
How machine learning enables wearable technology
In their experiments, the researchers tested the wearable technology on eight healthy adults. They collected data on the movement of the laryngeal muscles and used a machine learning algorithm to correlate the resulting signals with certain words. They then selected a corresponding output speech signal through the device’s actuation component.
The research team demonstrated the system’s accuracy by having participants speak five sentences, both aloud and silently, including “Hello, Rachel, how are you today?” and I love you!”
The overall prediction accuracy of the model was 94.68%, with the participants’ voice signal amplified by the performance component, demonstrating that the detection mechanism recognized their laryngeal movement signal and matched the corresponding sentence they the participants wanted to say.
In the future, the research team plans to continue expanding the device’s vocabulary using machine learning and test it on people with speech disorders.
Other authors of the paper are UCLA graduate students Samueli Ziyuan Che, Chrystal Duan, Xiao Wan, Jing Xu and Tianqi Zheng, all members of Chen’s lab.
The research was funded by the National Institutes of Health, the U.S. Office of Naval Research, the American Heart Association, the Brain and Behavior Research Foundation, the UCLA Clinical and Translational Sciences Institute, and the UCLA Samueli School of Engineering.