它不仅能有效地在语音识别中引入帧间相关信息,而且能克服状态输出概率密度函数为混合高斯分布的束缚。
The method not only would be effective to describe the inter-frame correlation information, but overcome tie of the state probability density function being mixture Gauss distribution.
而在各高斯分量概率密度互不重叠的条件下,使用动态簇算法(DC)则可快速而精确地估计出混合高斯模型参数。
And if there are no overlaps between each Gaussian component, parameters of Gaussian mixture PDF model can be exact estimated quickly with the dynamic cluster algorithm (DC).
应用推荐