摘要
文中介绍了一种基于FPGA的DTW说话人识别算法的实现方法,根据算法本身的特点,采用了一种适合于硬件实现的搜索路径约束条件,并利用收缩阵列来处理模板匹配中的运算,从而提高了算法的模板匹配速度和识别效率,在处理性能上优于现在常采用的基于软件的实现方法,适用于电话语音信道的实时大容量的说话人识别系统。
The paper describes the implementation of speaker recognition algorithms—dynamic timing wrapping based on FGPA.In the light of its characteristics, a restriction condition of searching path suitable for hardware is adopted, and the systolic array is used for the pattern-matching operation. The recogn-ition rate and speed is greatly improved. The method surpasses software implementation in performance and can be used in real-time recognition system of telephone network.
出处
《通信技术》
2008年第2期66-69,共4页
Communications Technology
