摘要 |
PROBLEM TO BE SOLVED: To improve operability and responsiveness when retrieving an institution name or the like by avoiding the execution of a troublesome procedure for following the hierarchical structure of a voice recognition dictionary having a hierarchical structure in order, and determining all conditions for narrowing down. SOLUTION: The voice recognition device is composed of a plurality of voice recognition dictionaries having a hierarchical structure (recognition dictionary storage part 105), an extraction means to extracts a desired dictionary as a waiting word list in the voice recognition dictionaries (control part 107), a selection means to select the operation of the desired dictionary in the voice recognition dictionaries (recognition dictionary selection part 104), a storage means to store the dictionary selected by the selection means as the waiting word list of the uppermost hierarchy with a usual dictionary extracted by the extracting means (RAM 103), and a recognition means to recognize an inputted voice by comparing the inputted voice with the waiting word list stored in the storage means (recognition part 102). |