Artificial intelligence device

1. An artificial intelligence device comprising:

a speaker;

a microphone configured to receive speech from a user; and

one or more controllers configured to:

extract an utterance feature of the received speech;

determine a user type corresponding to the extracted utterance feature;

map a speech agent associated with the determined user type; and

output an audio response through the speaker using the mapped speech agent,

wherein the utterance feature includes one or more of a word, a topic, a dialect, an accent, a stem, an ending, a pitch or an utterance speed of the received speech,

wherein the one or more controllers is further configured to:

extract features from a plurality of feature items included in the utterance feature; and

delete one or more features based on a degree of importance from the extracted features,

wherein the determined user type is categorized by features remaining after the deletion of the one or more features.

2. The artificial intelligence device of claim 1 , wherein the one or more controllers is further configured to place the user type and the speech agent in a feature space and map a user type and a speech agent located at a nearest position in the feature space.

3. The artificial intelligence device of claim 1 , wherein, the one or more controllers is further configured to map the user type to the speech agent for outputting a speech with a dialect of a specific area based on the utterance feature including the dialect of the specific area.

4. The artificial intelligence device of claim 1 , wherein, the one or more controllers is configured to map the user type to a speech agent for outputting a speech at a specific utterance speed based on the utterance feature including the specific utterance speed.

5. The artificial intelligence device of claim 1 , wherein, the one or more controllers is further configured to map the user type to a speech agent for outputting a speech with a dialect of a specific area at a specific utterance speed based on the utterance feature including the dialect of the specific area and the specific utterance speed.

6. An artificial intelligence device comprising:

a speaker;

a microphone configured to receive a first speech from a user; and

a controller configured to:

extract a first utterance feature of the received first speech;

determine a first user type corresponding to the extracted first utterance feature;

map a first speech agent associated with the determined first user type;

output a first audio response through the speaker using the mapped first speech agent;

receive a second speech; and

map a second speech agent associated with a second user type corresponding to a second utterance feature extracted from the second speech based on the second utterance feature being changed from the extracted first utterance feature.

7. The artificial intelligence device of claim 6 , wherein the controller is further configured to:

extract the second utterance feature from the second speech;

determine the second user type corresponding to the extracted second utterance feature based on the extracted second utterance feature being different from the first utterance feature; and

output a second audio response through the speaker using the mapped second speech agent.

8. The artificial intelligence device of claim 7 , wherein the first utterance feature includes one or more of a word, a topic, a dialect, an accent, a stem, an ending, a pitch or an utterance speed of the first speech.

9. The artificial intelligence device of claim 8 , wherein, the one or more controllers is further configured to select the second speech agent for outputting responses at a second utterance speed when the first utterance feature is a first utterance speed and the second utterance feature is a second utterance speed.

10. The artificial intelligence device of claim 8 , wherein the one or more controllers is further configured to:

extract features from a plurality of feature items included in the first utterance feature; and

delete one or more features with a low degree of importance from the extracted features,

wherein the determined first user type is categorized by features remaining after the deletion of the one or more features.

11. The artificial intelligence device of claim 10 , wherein the one or more controllers is further configured to place the user type and the speech agent in a feature space and map a user type and a speech agent located at a nearest position in the feature space.