小波包变换频带划分在MELP声码器中的应用  

Frequency Band Division of MELP Vocoder Using Wavelet Packet Transform

作　　者：许钢[1] 黄冰[1] 

机构地区：[1]桂林电子工业学院通信与信息工程系,广西桂林541004

出　　处：《电声技术》2004年第2期37-41,共5页Audio Engineering

摘　　要：提出了一种基于小波包变换划分信号频带的方法,并以此为基础对MELP声码器基音检测算法做出改进。原始语音信号经过多尺度正交小波包变换得到MELP声码器所要求的多种子带信号输出;然后对0～1kHz和0～500Hz子带输出信号进行基音周期的估计。实验表明,该方法提取出的基音周期不仅具有较高的可靠性和精度,而且对环境噪声有较强的稳健性。A signal frequency band division algorithm based on wavelet packet transform is presented in this paper. On the basis of that, the pitch detection algorithm of MELP vocoder is improved greatly. Many sub-bands signals are obtained through multi-scale orthogonal wavelet packet transform of the original speech, whose pitch period is estimated separately according to 0～1 kHz and 0～500 Hz sub-bands signals. The experiment results show that this algorithm has not only high reliability and precision, but also robustness againsst the environmental noise.

关 键 词：小波包变换 频带划分 MELP声码器 基音检测 混合激励线性预测 

分 类 号：TN912.32[电子电信—通信与信息系统]