摘要 |
PROBLEM TO BE SOLVED: To prevent impersonation with a reproduced voice and identify an individual from an input voice.SOLUTION: A voice identification device comprises: an input part 12 which inputs an acoustic signal including an audible band and an ultrasonic band; an individual identification part 15 which identifies an individual from the acoustic signal input from the input part 12; a band extraction part 13 which extracts the signal of the ultrasonic band from the acoustic signal input from the input part 12; and a live voice detection part 16 which determines whether or not the acoustic signal input from the input part 12 is a live voice which is input live by using the extracted signal of the ultrasonic band. |