PROBLEM TO BE SOLVED: To detect a user's utterance in a dialogue with an unknown user. SOLUTION: A computer outputs an utterance request requesting a user to utter a predetermined word, and performs voice recognition for a first voice acquired after the utterance request is output, thereby performing a voice recognition of the first voice. Generate recognition results. When the recognition result of the first voice is similar to the predetermined word, the computer registers the feature information of the first voice in the storage unit, and the first voice is acquired based on the feature information of the first voice. From the second voice acquired later, the voice whose speaker is the user is detected. [Selection diagram] FIG. 4
展开▼