In the state of recognition, after calculating the character vector of input voice, we compare it with the character vectors of templates, and then find the best one or refuse it.

识别时，计算出输入语音的特征矢量参数，依据模板匹配的方法与模板库中的语音模板逐一进行比较，得出最佳的匹配模板，或作出拒识判断。

youdao

It picks up the character vectors of a group of Numbers from the sequence of the character vectors of the input speech signals, and then put the vectors into the NN sorter to recognize.

该网络从输入语音信号的特征矢量序列中提取出一组固定数目的特征矢量，然后将这特征矢量馈入神经网络分类器进行识别。

youdao

After the training, characters are extracted from the2 0 unclassified artificial sequence samples and1 82 natural sequence samples to form the character vectors as input of the two NN for clustering.

通过训练后，将2 0个未分类的人工序列样本和1 82个自然序列样本提取特征形成特征向量并输入两个网络进行分类。

youdao