摘要 |
<p>PURPOSE: To improve operability and to facilitate re-input of a voice at the time of erroneous recognition by attaining a voice input only when one button is depressed and performing the recognition of an input voice only when the input is specified one. CONSTITUTION: When an input button 1 is operated, a voice input part 3 is controlled so that the voice input from a microphone 3a is validated. For instance, when the voice input '3' exists, the recognition corresponding to that is performed by a recognition part 4 from a voice pattern stored in a memory 4a. When the recognition processing is performed, the recognition result is transmitted from the recognition part 4 to an information part 7. The information part 7 makes a user discriminate the correct/error of the recognition by the display, etc., of the data based on the recognition result. In such a case, e.g. it is assumed that '7' is recognized corresponding to the voice input '1'. Since that is erroneous recognition, the user releases the operation of the input button 1 temporarily, and thereafter, the voice input is performed again in a prescribed time, and then correction is performed.</p> |