基于DSP的语音时延估计实现

Implementation of Voice Time-delay Estimation Based on DSP

基于麦克风阵列的语音时延估计是语音增强和空间定位的关键技术,实现了对时延估计的几种算法做了仿真和比较,研究了利用数字信号处理器(DSP)芯片来实现时间延时估计的软硬件实现,提出并实现了基于TI公司的TMS320C6416芯片和TLV320AIC23芯片的两路语音信号采集处理。并基于语音信号的特殊性,提出了计算一段时间内时延值的特殊方法。仿真和实验的结果表明提出的方法能够较准确的计算和检测出一段时间内两路语音信号的时间延时。

Voice delay estimation based on microphone array is the crucial technology in the area of voice enhancement and spatial orientation.Simulation and comparison on several delay estimation algorithms are done.Moreover,the hardware and software implementations for delay estimation by DSP are explored,and the two-way voice signals acquisition and processing based on TMS320C6416 and TLV320AIC23 is proposed and impleniented.According to the particularity of voice signal,a special method for computing a delay value in a period of time is given.Simulation and experimental results show that the proposed method could calculate and detect accurately the delay of two-way voice signal in a period of time.