| 摘要 |
In a method of recognizing an input pattern represented by a time sequence of input feature vectors (a(t)) as a recognized pattern selected from a plurality of reference patterns which represent categories of recognition objects, respectively, each of the reference patterns is defined by a sequence of state models, successively supplied with the time sequence of the input feature vectors and with a sequence of preceding state vectors (h(t, s, n)), the sequence of the state models produces a time sequence of predicted feature vectors (A(t+l, s, n)) and a sequence of new state vectors (h(t+l, s, n)). The recognized pattern is selected from one of the reference patterns that minimizes a prediction error between the time sequence of the input feature vectors and the time sequence of the predicted feature vectors. The prediction error is calculated by using a dynamic programming algorithm. Training of the reference pattern is carried out by a gradient descent method such as back-propagation technique.
|
