一种改进的皮肤听声语音信号处理系统

An improved speech processing system to skin-hearing

针对皮肤听声器模拟系统中相邻通道通带重叠度高、刺激电流频率不一致等问题,使用LabVIEW软件将语音预加重后进行短时傅里叶变换,根据划分的频带范围计算各通道的能量和,对能量和进行非线性压缩处理后调制输出,实现了改进的皮肤听声语音信号处理系统.通过单频信号和语音信号对系统进行了测试,结果表明系统运行稳定,各通道输出正确,输出信号频率固定.相比传统的皮肤听声器,系统修改灵活、测试方便,减少了各频带的频率对输出的影响,增大了各通道输出的差异,对皮肤听声技术的发展具有非常重要的指导意义和研究价值.

Aiming at the problems of adjacent channel passband overlap and the stimulus current frequency consistency for analog system of skin-hearing aid,we used the software of LabVIEW to pre-accentat the speech and then carry out the short-time Fourier transform. The energy of each channel was calculated according to the range of frequency band that had been divided. The signals of energy were modulated and output after nonlinear compression. And then an improved speech processing system of skin-hearing was achieved. The system was tested by the single-frequency signal and voice signal. The experimental results showed that the system was stable and the outputs were corrected in each channel. In addition,the frequencies of the output signal were stationary.Compared with the traditional skin-hearing aid,this system is easier to modify,more convenient for testing. It also reduces the impact of frequencies on each channel and increases differences of each channel. This system has significant meaning and research value for the development of skin-hearing technology.