High-Sensitive, Wireless, Real-Time Throat Microphone for Communication in Noisy Environment

Abstract

Voice is the most significant interface in the Internet of Things (IoT), such as smart factories and smart cites. People can control the devices through voice commands while working. However, it is difficult to transmit voice as ambient noise increases. Therefore, a reliable communication method is required to form a stable interface between IoT devices. The throat microphone attached onto the neck skin can detect voice in noisy environment. However, previous throat microphones are still insufficient to represent voice accurately due to their rigid form factor and internal noise. Here, we introduce a noise-cancelling wearable throat microphone with a flexible material and lightweight. Signal quality of the throat microphone is improved by removing the internal white noise through an on-device modified 1-D Kalman filter. The proposed filter increased the segmental signal-to-noise ratio (SNR) from 26.36 dB to 45.23 dB. In addition, the throat microphone attachment method was optimized by comparing the quantitative difference between the throat microphone signal and the reference microphone signal. The optimal attachment position was the upper part of the thyroid cartilage, and the optimal covering pressure was evaluated to be 4 mmHg. Finally, we demonstrated the potential use of a wearable throat microphone in wearing gas mask and various environments such as noisy and windy.