一种改进的基于小波变换的语音基频检测  

An Advanced Pitch Detection Algorithm Based on Wavelet Transform

作　　者：张国伟[1] 田岚[1] 

机构地区：[1]山东大学信息科学与工程学院,山东济南250100

出　　处：《临沂师范学院学报》2006年第6期101-105,共5页Journal of Linyi Teachers' College

摘　　要：传统的小波变换基频检测通过比较相邻尺度上的小波系数极值点来进行检测．该方法往往存在着伪极值点的误判,降低了检测的精确度,而且增加了运算量．针对小波系数的伪极值点误判情况做了分析,进行了最佳尺度的判定,以及极值点搜索方法的改进,提出了一种基于最佳尺度的小波变换基频检测算法,提高了算法的实时性和准确性,并通过实验证明了该算法的有效性．The traditional pitch detection algorithm based on wavelet transform extracts the fundamental frequency by comparing the positions of maximum wavelet coefficients of continuous scales. This strategy may take false maximum points as real ones, and reduces the accuracy of the algorithm. By researching the false maximum wavelet coefficient points, this paper proposes an advanced algorithm in the basis of optimum scale and a new peaksearching method. The validity of the algorithm is testified by experiments.

关 键 词：基频检测 小波变换 最佳尺度 

分 类 号：TN912.34[电子电信—通信与信息系统]