摘要 |
PROBLEM TO BE SOLVED: To achieve a higher operability of a voice recognizing device having a recognition system especially hierarchized concerning a voice recognition method for recognizing voice and an apparatus for implementing it in an information processor such as computer and a car navigation system. SOLUTION: Some of words as candidate below the second candidate position obtained as a result of the first voice recognition are selected to be added to the second voice recognition word to simplify recovery operation in erroneous recognition. Even when erroneous recognition occurs in the first recognition operation, the uttering of speech of the word the same as the first one is repeated by the second operation to allow the reselection of the first word again. This enables reselection of the word to be selected again without cancelling operation. Thus, a user can use the hierarchy of a voice recognizing device without consciousness. |