在此基础上，建立了最小方差损失函数，并结合高斯·牛顿预测误差方法，提出了稳定的，高性能的，在线的复频率直接估计算法。
A cost function is presented, and by applying Gaussian-Newton type recursive prediction error based method, a stable and efficient online frequency estimation algorithm is derived.
在此提出一种利用线性预测误差去除语音中的加性白噪声的方法。
It presents a new method on reducing additive white noise in speech signal using linear prediction error.
由于它考虑了几种方法所获得结果的权重，因此可减小预测误差，提高预测的准确性，是一种值得推广的预测方法。
It can reduce the forecasting error, improve the forecasting accuracy as it has considered the weights obtained by several methods. Therefore, this forecasting method is worth spreading.
应用推荐