| 摘要 |
PURPOSE:To allow the control of a machine with such sensation that human beings carry on conversation with each other by providing a controlled variable setting section which determines a controlled variable by at least two of the result of the recognition by a voice recognizing section, the result of classification by a vocabulary classifying section and the controlled variable stored in a memory. CONSTITUTION:The word which is the information on the result of the recognition is classified to the word belonging to either of the basic word or auxiliary word in the vocabulary classifying section 3. The control by as much as the value for the recognized word is executed in the 1st controlled variable setting section 4 and this quantity is stored in a memory 6 in the case of the basic word. If the result of the recognition is the auxiliary word, the control is executed when the word is sent from the vocabulary classifying section 3 to the 2nd controlled variable setting section 5. The controlled variable is also stored in the memory 6. The control section 7 controls an appliance 8 to be controlled by the controlled variable. If the quantity controlled immediately before is previously stored into the memory, meaning can be provided in a adverb even if the adverb is next inputted by a voice. Since the voice input can be made as if the conversions are carried out between people and, therefore, the laborious dealing with the voice controller is relieved and the controlled operation is executed in the sensation near to conversation with people. |
