声频信号特征波形的合成方法

Method for composing characteristic waveform of audio signals

Abstract

声频信号特征波形合成及选取方法,是先对待处理的波形进行分析,筛选出具代表性的特征波形。在存储时,仅记录这些特征波形及其相关参数,不必记录整个信号的波形,可减省大量的存储空间。在后续合成还原时,通过读出此特征波形及其相关参数,再以一内插运算法就可以合成还原出波形。其所合成的音质接近自适应差分脉码调制(ADPCM)的音质,故符合低速中央处理器的应用领域。
A method for composing and choosing the characteristic waveform of audio signals includes analyzing the waveform to be processed, choosing representative characteristic waveforms, only storing said characteristic wave forms and relevant parameters for saving a lot of memory space, and composing and reproducing the waveform by reading out the characteristic waveforms and relevant parameters and using interpolation algorithm. The resultant sound has high quality comparable to that of adaptive differential pulse code modulation (ADPCM), so meeting the application domain of low-speed CPU.

Claims

Description

Topics

Download Full PDF Version (Non-Commercial Use)

Patent Citations (0)

    Publication numberPublication dateAssigneeTitle

NO-Patent Citations (0)

    Title

Cited By (3)

    Publication numberPublication dateAssigneeTitle
    CN-100421153-CSeptember 24, 2008顾稚敏;宋 明一种预存式语言识别系统及其方法
    CN-101710488-BAugust 03, 2011安徽科大讯飞信息科技股份有限公司语音合成方法及装置
    CN-103903610-BFebruary 08, 2017北京谊安医疗系统股份有限公司一种制作合成音频数据的方法