Tokyo, February 27, 2018 - NEC Corporation (NEC; TSE: 6701) today announced the development of a biometric recognition technology that identifies individuals by measuring the differences in the shapes of ear cavities with sound that is inaudible to the human ear. This enhanced ear acoustic authentication technology (Note 1), which is expected to be available by the end of 2018, was developed in cooperation with Nagaoka National College of Technology.

Conventional ear acoustic authentication is audible to the human ear and tends to cause discomfort that interferes with work, concentration and hearing ambient sound. This new technology reinforces security through full-time (constant) recognition that does not interfere with the behavior or work of users.

The technology accurately identifies individuals by measuring acoustic characteristics that represent the shape of a user's ear canal. Smart earphones, which are also equipped with microphones, emit high-frequency (18 to 48 kHz) inaudible sound, which is then reflected from the ear canal (Figs. 1 and 2) and captured by the microphones. Analysis of the reflected sounds enables the characteristics of an individual's ear canal to be determined.

"NEC developed this technology with a view to preventing identity fraud and ensuring safety and confidence in the maintenance, management, and security of important social infrastructure," said Akio Yamada, general manager, Data Science Laboratories, NEC Corporation. "This technology can help to increase the efficiency of operations at medical facilities and call centers, for example, through hands-free recognition that is provided through the use of smart earphones, which users are already comfortable with using in daily life."

Fig. 1: Audible field for select animals (audible frequency range). NEC's new technology uses sound of 18 to 48 kHz (wavelength: 7 to 18 mm) that is inaudible to humans (Note 2)

Conventional technology

New technology

Fig. 2: Sound that is inaudible to the human ear is sent, and individuals are identified by the reflected sound

Fig. 3: The differences in the shape of the ear canal, which is unique for each person, can be measured using audible sound and inaudible sound