| 摘要 |
PURPOSE:To decrease the error occurrence rate of correction, etc., by voice input by completing the registration of one instruction word and registering only the voice of a specific speaker when the same voice signal is voiced twice continuously at the time of registering the voice signal. CONSTITUTION:A voice signal storage means 52 converts a voice signal inputted to a microphone into a digital signal and stores it in word areas of a register and a storage element, and outputs the address signal of the word area. A 1st register 28 stores the address of the word area where the voice signal is registered. A 2nd register 30 stores the registration address of the word area where the same signal as the input voice signal is registered. A coincidence detecting means 32 outputs a coincidence signal when the contents of the 1st register 28 and the 2nd register 30 become equal to each other. A registration address specifying means 34 outputs the address specification signal of the word area, and receives the coincidence signal from the means 32 to update the registration address of the word area. When the voice signal is registered, the registration of the instruction word is not completed unless the same voice signal is voiced twice successively, and only the voicing of the specific speaker is registered in the word area. |
