对采样过后的语音信号进行分帧处理，并将分帧后的信号存储为一个矩阵-right after sampling the voice signal processing sub-fr a me and sub-fr a me of a signal to the storage matrix