中文说明： 语音信号是时变。单个参数信号的改变比信号本身较缓慢。因此，对于此类参数的测量，显著较低的频率是信号的 ↗sampling 比 ↗sampling 频率的必要条件。帮助下窗函数的加权信号在时间域和分为局部信号序列进行了测量。关于测量的确切目标，必须确定"短时间分析"两个维度： 窗口长度 （加权部分信号长度） 和测量时间间隔 （帧速率，即各个窗口之间的距离）。 短时间分析操作 短时能量： 它是语音信号的强度。它反映了语音信号的振幅变化。 短时自相关函数： 自相关函数用于在函数中查找的周期性和 是很多的频谱分析方法的基础。 短时过零率： 短时过零率被定义为 次数的语音信号变化签署内给定 时间窗口。
English Description：
The speech signal is time-varying. The change of single parameter signal is slower than the signal itself. Therefore, for the measurement of such parameters, a significantly lower frequency is necessary ↗ Sampling ratio ↗ The necessary condition of sampling frequency. With the help of the window function, the weighted signal is measured in time domain and divided into local signal sequences. As for the exact target of measurement, two dimensions of "short time analysis" must be determined: window length (weighted part signal length) and measurement interval (frame rate, that is, the distance between windows).