Professors at POSTECH Develop a Vibration Sensor to Recognize Voice Despite Surroundings


Image Source: Soda PDF

A research group from South Korea’s Pohang University of Science and Technology (POSTECH) developed a vibration sensor that attaches to the neck and recognizes voice through air vibration. POSTECH claims that the precision of recognition is not affected by ambient noise or volume, unlike mobile phones that use a microphone to detect sound. Common voice activation concerns, like a device not detecting a command or an assistant activating unprovoked, could be prevented with this responsive sensor thanks to Professors Kilwon Cho and Yoonyoung Chung who led the development.

“This research is very meaningful in a way that it developed a new voice-recognition system which can quantitively  sense and analyze voice and is not affected by the surroundings. It took a step forward from the conventional voice-recognition system that could only recognize voice qualitatively, said Cho of the study”

Wearable healthcare monitoring device

POSTECH described applicable use cases for the sensor’s technology, including healthcare, which remains one of the growing markets for AI and voice assistants. “This research can be further extended to various voice-recognition applications such as an electronic skin, human-machine interface, wearable vocal healthcare monitoring device.”

Amazon is also testing a wearable emotion recognition gadget. A few months ago it was learned the Alexa voice software team is working on a health and wellness device that is to be worn on the wrist to detect the emotional state of the wearer based on the inflection of their voice.

Researchers from The University of Washington also announced a tool last month that detects cardiac arrest. Their system is contactless and will function through a skill or smart speaker, however it uses the same methodology of accurately depicting noises for the purpose of medical prevention or detection.

Wearables & Services are the Future of Apple

Amazon Testing Emotion Recognition Gadget