论文部分内容阅读
题目二 语音识别 有一个研究小组用计算机处理语音,他们首先将语音电信号用如下办法转换为机内表示:在等间隔的时间里依次对语音信号的电平进行测试,如果为高电平则记录1,为低电平则记录0,这样就得到了一个输入信号的由0、1组成的字串,该方法称为采样。为了处理方便起见,在采样时将字串的长度限定为2~k(k=0,1,…)。
Topic Two Speech Recognition A team of researchers used computers to process speech. They first converted speech signals into speech signals in the following ways: testing the speech signal levels in equal time intervals; if high, Record 1, record low 0, so that you get an input signal composed of 0,1 string, the method is called sampling. For processing convenience, the length of the string is limited to 2 ~ k (k = 0, 1, ...) during sampling.