The proposed method uses the maximum singular value of an cumulant matrix to distinguish between voiced parts of the speech signal and noise.
提出一种采用累积量矩阵的最大奇异值来实现语音端点检测的方法，并引入一种自适应的实现方法。
The proposed method uses the maximum singular value of an cumulant matrix to distinguish between voiced parts of the speech signal and noise.
提出一种采用累积量矩阵的最大奇异值来实现语音端点检测的方法，并引入一种自适应的实现方法。
应用推荐