基于线性预测分析的语音三维语图绘图方法  

A Method of Drawing Speech 3-D Spectrogram Based on Linear Prediction Analysis

作　　者：田岚[1] 白树忠[2] 南新志 

机构地区：[1]山东工业大学电子系,济南250061 [2]山东工业大学电力系

出　　处：《控制与决策》1997年第4期365-368,共4页Control and Decision

基　　金：山东省自然科学基金资助课题

摘　　要：介绍在通用PC机上实现语音三维语谱绘图软件的设计方法。根据线性预测分析技术,采用两种出图方式,通过软件编程将提取的频谱包络在显示器和打印机上绘制成图。其绘图功能及效果可与绘图仪相比拟,具有方便灵活的特点。A programming design which can be used for drawing speech 3-D spectrogram on PC computer is introduced. Based on linear prediction analysis, speech spectrum contours can be extracted and drawed out in different ways, such as on displayer and printer. This method has similar effects as by spectrograph. It is easy and flexible to operate or control.

关 键 词：线性预测分析 三维语图 语音信号处理 语图仪 

分 类 号：TN912.3[电子电信—通信与信息系统] TP391.41[电子电信—信息与通信工程]