Air- and bone-conductive integrated microphones for robust speech detection and enhancement.pdf

AIR- AND BONE-CONDUCTIVE INTEGRATED MICROPHONES FOR ROBUST SPEECH DETECTION AND ENHANCEMENT Yanli Zheng?, Zicheng Liu, Zhengyou Zhang, Mike Sinclair, Jasha Droppo, Li Deng, Alex Acero, Xuedong Huang? Microsoft Research Redmond, WA 98052 ABSTRACT We present a novel hardware device that combines a reg- ular microphone with a bone-conductive microphone. The device looks like a regular headset and it can be plugged into any machine with a USB port. The bone-conductive microphone has an interesting property: it is insensitive to ambient noise and captures the low frequency portion of the speech signals. Thanks to the signals from the bone- conductive microphone, we are able to detect very robustly whether the speaker is talking, eliminating more than 90% of background speech. Furthermore, by combining both chan- nels, we are able to significantly remove background speech even when the background speaker speaks at the same time as the speaker wearing the headset. 1. INTRODUCTION One of the most difficult problems for an automatic speech recognition system is in dealing with noises. When there are multiple people speaking, it is difficult to determine whether the captured audio signal is from the speaker or from other people. In addition, the recognition error is much larger when the speech is overlapped with other people’s speech. Because the speech is non-stationary, it is extremely hard to remove the background speech from just one channel of audio signals. In this paper, we propose a hardware device that com- bines regular microphone (air-conductive microphone) with a bone-conductive microphone with the purpose of handling noisy environment. The device is designed in such a way that people wear it just like a regular headset, and it can be plugged into any machine with a USB port. Compared to the regular microphone, the bone-conductive microphone is insensitive to ambient noise but it only captures the low frequency portion of the speech signals. Because it is insen- s