Identifying user of voice operation based on voice information, voice quality model, and auxiliary information

A voice operation system comprising:

• an array microphone comprising a plurality of microphones that each detects voice vibration and converts the voice vibration into voice information of a person who is speaking at a current time;
• a storage medium that stores registered user information in which:
• first user information registers a direction of an expected location of a first user and with a first time period in which the first user is expected to be speaking;
• and
• second user information registers a direction of an expected location of a second user and with a second time period in which the second user is expected to be speaking;
• a memory storing instructions;
• and
• a processor configured to execute the instructions to:
• perform voice recognition on the voice information;
• estimate an arrival direction of the voice vibration as detected by the array microphone;
• identify that the person who is speaking at the current time is the first user in response to the current time matching the first time period and the arrival direction matching the direction of the expected location of the first user based on the first user information and the estimated arrival direction;
• and
• identify that the person who is speaking at the current time is the second user in response to the current time matching the second time period and the arrival direction matching the direction of the expected location of the second user based on the second user information and the estimated arrival direction.