"Hey, Siri, how is the weather today?" speech recognition technology is becoming increasingly popular. It is a convenient technology that is widely used. However, in order to fully utilize its intended functionality, the user must stand near the device and pronounce the words carefully. What if the skin on our bodies could recognize sounds without a device?
Professor Kilwon Cho and Dr. Siyoung Lee of the Department of Chemical Engineering, and Professor Wonkyu Moon and Dr. Junsoo Kim of the Department of Mechanical Engineering at POSTECH developed a microphone to detect sound by applying polymer materials to microelectromechanical systems (MEMS)
This compact microphone displays a wider hearing field than the human ear while attaching easily to the skin. The academic results were recently published in the journal Advanced Materials.
Traditional MEMS microphones used in mobile phones, Bluetooth devices and other devices are composed of thin, small and complex diaphragm structures. However, because they are made of hard, brittle silicone, it is difficult to bend the diaphragm or microphone as needed, which can interfere with the device's sound detection capabilities. The
research team overcame this limitation and created a MEMS-based microphone structure using a polymer material that is more flexible than silicon and can be designed into any shape. The device is a quarter the size of a fingernail and only a few hundred micrometers (μm, 1μm = one millionth of a meter) thick. The microphone can be mounted on a large surface area of the body or even on a finger.
According to research, the microphone's hearing sensitivity is higher than that of the human ear, while recognizing surrounding sounds and the user's voice without distortion. In addition, it can detect loud sounds exceeding 85 decibels (the range that causes hearing damage) and low-frequency sounds that are inaudible to humans.
The quality of speech detection is comparable to that of a cell phone or recording studio microphone. When the acoustic sensors on the skin are connected to a commercial voice assistant program ( Google Assistant ), users can search, translate and control the device effortlessly.
This new acoustic sensor has potential applications in wearable speech recognition devices for the Internet of Things and human-machine interfaces. The research team plans to create auditory electronic skin by integrating it with skin-connectable pressure and temperature sensors, flexible displays, , etc.
