摘要
该文主要叙述数字语音信号的基于FFT的非均匀采样的技术实现问题,涉及WAV文件格式、分段、对其进行非均匀降速重采样频率的选择原则及生成和语音信号重构,并对实际的WAV格式的数字语音文件用VC++6.0编写的程序实现了基于FFT的重采样;分析方法、实现程序不仅对WAV格式的数字语音文件有效,而且也适用于其它格式的数字语音文件和非语音信号的非均匀采样的实现。
Implementation of nonuniform sampling and recovery of speech signal stored in WAV format is described in the paper.In the resampling,digital speech signal is equivalently segmented and then different re-sampling frequency,which is automatically determined by using Discrete fast Fourier transform(FFT),is used according to frequency characteristics of each segment.The implementation methods and software programs are not only effective to digital speech signals of WAV format,but they can be extended to other formats also.Moreover,it is helpful to nonunifrom resampling of digital signals or sequences than other speeches.
出处
《计算机工程与应用》
CSCD
北大核心
2002年第15期31-32,42,共3页
Computer Engineering and Applications
基金
江苏省自然科学基金资助
