| 摘要 |
<p>PROBLEM TO BE SOLVED: To provide a method and voice input person decision device which can securely decide a voice input person without requesting the decided person to do troublesome operation. SOLUTION: The voice input person decision device 11 decides who is a human in the scene where the human speaks to a communication robot. For the purpose, the device is equipped with a voice detecting means 21 which detects a sound inputted from a microphone 1, a 1st database 22 in which voice signal characteristics of a plurality of humans are registered, a 2nd database 23 in which pieces of individual identification information on the plurality of humans are registered, a voice recognition processing means 24 which matches signal characteristics of the sound detected by the sound detecting means 21 against respective voice signal characteristics registered in the 1st database 22 to estimate the person who has highly matching voice signal characteristics as the voice input person, a voice input person confirmation processing means 25 which takes the individual identification information on the person out of the 2nd database 23 and outputs it through an LCD 2 and a speaker 3, and a final determination processing means 27 which determines the voice input person when there is a specific answer to the outputted individual identification information.</p> |
