| 摘要 |
<p>A continuous speech recognition system comprises means for developing a similarity measure and a path information at each point (i, j) in the order of monotonous increase of i and j based on the similarity measure and path information obtained up to said time point (i, j) by making use of Dynamic Programming, i and j designating time points at time axes of an input pattern and a continuous reference pattern, respectively, the input pattern being given in a time series of feature vectors for input speech, the continuous reference pattern being given in a time series of feature vectors for connected combinations of a plurality of reference word patterns, the similarity measure being given by an accummulated distance between the feature vectors of the input pattern and the reference word patterns, and the path information representing a start time point of a word unit of the continuous reference pattern; means for setting the minimum similarity measure obtained for a word unit as an initial value for a next subsequent word unit; means for checking whether or not the path information at the time point (i, j) is within a predetermined window restriction; means for excluding a time point outside the window restriction from the development of the similarity measure and path information; and means for deciding recognition results of the input speech on the basis of the matching path showing the minimum similarity measure at the end time point of the input pattern.</p> |
