Electronic apparatus and control method thereof

A method of controlling an electronic apparatus comprising a microphone and a processor, the method comprising:

• receiving a user voice input by the microphone;
• in response to the user voice being input through the microphone, calculating, by the processor, a plurality of similarities between a voice print feature of the user voice and a plurality of voice print features of a plurality of voice print feature models representing a plurality of user voices without using, in calculating the plurality of similarities, an utterance environment model corresponding to a place where the user voice is input;
• determining, by the processor, whether there is a similarity equal to or greater than a preset value, from among the plurality of similarities;
• in response to determining that the similarity equal to or greater than the preset value exists, recognizing a user, by the processor, based on a voice print feature model corresponding to the similarity equal to or greater than the preset value, from among the plurality of voice print feature models;
• and
• in response to determining that the similarity equal to or greater than the preset value does not exist, revising, by the processor, the plurality of similarities using the utterance environment model corresponding to the place where the user voice is input, from among a plurality of utterance environment models representing a plurality of environmental disturbances, and recognizing the user, by the processor, based on the revised plurality of similarities.

An electronic apparatus, comprising:

• a microphone configured to receive a user voice;
• a processor, in response to the user voice being input through the microphone, configured to:
• extract a voice print feature corresponding to the user voice,
• extract surrounding noise information incorporated in the received user voice using a surrounding noise information model stored in a storage unit,
• calculate a plurality of similarities between the voice print feature of the user voice and a plurality of voice print features of a plurality of voice print feature models which represent a plurality of users and are stored in the storage unit, without using, in calculating the plurality of similarities, the surrounding noise information,
• determine whether there is a similarity equal to or greater than a preset value, from among the plurality of similarities,
• in response to determining that the similarity equal to or greater than the preset value exists, recognize a user based on a voice print feature model corresponding to the similarity equal to or greater than the preset value, from among the plurality of voice print feature models, and
• in response to determining that the similarity equal to or greater than the preset value does not exist, revise the plurality of similarities using the surrounding noise information, and recognize the user using the revised plurality of similarities.