Burshtein D
Tel-Aviv University, Department of Electrical Engineering-Systems, Israel.
J Acoust Soc Am. 1992 Mar;91(3):1531-7. doi: 10.1121/1.402485.
The well-known speech production model is considered, where the speech signal is modeled as the output of an all-pole filter driven either by some white noise sequence (unvoiced speech) or by the sum of a periodic excitation and a noise sequence (voiced speech). Approximate maximum-likelihood (ML) estimation algorithms for the unvoiced case are well known. The ML estimator of the parameters is obtained for the voiced speech model. These parameters consist of the parameters of the periodic excitation (pitch parameters) and the parameters of the filter [linear prediction coefficient (LPC) parameters]. The results of the application of the algorithm on simulated and on real speech data are presented.
考虑了著名的语音生成模型，其中语音信号被建模为全极点滤波器的输出，该滤波器由一些白噪声序列（清音语音）或由周期性激励与噪声序列之和（浊音语音）驱动。清音情况下的近似最大似然（ML）估计算法是众所周知的。针对浊音语音模型获得了参数的ML估计器。这些参数包括周期性激励的参数（基音参数）和滤波器的参数[线性预测系数（LPC）参数]。给出了该算法在模拟语音数据和真实语音数据上的应用结果。
