Method and system for context association and personalization using a wake-word in virtual personal assistants

1. A method for generating a personalized response from a virtual assistant, the method comprising:

receiving a user query including a wake-word;

replacing unstructured inputs included in the user query with a machine-comprehendible structured form;

separating the wake-word and a remaining portion of the user query;

sending the wake-word to a wake-word processor;

parsing the remaining portion of the user query separately from the wake-word;

retrieving a query response from at least one knowledge base based on the wake-word;

generating a standard natural language response from the query response; and

synthesizing a personalized response from the standard natural language response based on the wake-word,

wherein retrieving the query response further comprises:

receiving a plurality of action steps from an action planner module;

searching for an appropriate knowledge base from the user query;

extracting the query response to the user query from the searched appropriate knowledge base;

instructing the query response to be sent to a response path planner module;

searching the user query in a plurality of knowledge bases;

preparing a routing plan for searching the query response in one of the plurality of knowledge bases; and

routing the user query to the appropriate knowledge base for retrieving the query response.

2. The method of claim 1 , further comprising:

extracting wake-word related information from a wake-word database;

parsing the extracted wake-word related information along with a plurality of user preference information;

classifying information received from a wake-word parser; and

producing a wake-word context from the classified information.

3. The method of claim 2 , further comprising:

generating a user query context from the wake-word context and a separated user query; and

computing the query response based on the user query context.

4. The method of claim 2 , further comprising assigning a plurality of tags to information received from the wake-word database.

5. The method of claim 2 , further comprising:

mapping the wake-word and the wake-word context; and

prompting a user to choose an appropriate wake-word when the wake-word and the wake-word context are not mapped.

6. The method of claim 1 , further comprising:

inputting a plurality of wake-words along with a plurality of wake-word parameters, wherein the plurality of wake-word parameters are set based on a plurality of user preference information including, at least one of, a response type, a voice of a response, a language, an avatar, or a filler source;

creating an entry for the plurality of wake-words along with the plurality of wake-word parameters and a plurality of knowledge base information in a wake-word database, wherein the plurality knowledge base information includes, at least one of, a profile identification (ID), the wake-word, or the knowledge base;

mapping the plurality of wake-words along with the plurality of knowledge bases;

storing the mapped plurality of wake-words with the plurality of knowledge bases in data storage at a server.

7. The method of claim 1 , wherein synthesizing the personalized response further comprises:

interacting with a server or a wake-word processor for extracting user preference information related to the wake-word;

receiving the user preference information;

converting the received user preference information along with a plurality of semantic fillers into a personalized response;

generating the personalized response to speech in, at least one of, a desired voice, action, form, or text; and

preparing the plurality of semantic fillers and choosing a user's preferred words for outputting the personalized response.

8. The method of claim 1 , further comprising:

extracting wake-word information from a server through a plurality of application program interface (API) calls; and

returning the wake-word information and user preference information to a wake-word parser in response to one of the plurality of API calls,

wherein a data structure of the wake-word information and the user preference information is a standard data format.

9. The method of claim 1 , wherein a plurality of wake-words are assigned corresponding to at least one mode from among a plurality of modes,

wherein the plurality of modes include a general mode and a private mode, and

wherein the private mode includes:

initializing the private mode based on a user query context;

changing at least one of a privacy setting, a response setting or a response type;

determining private information of a user and excluding the private information of the user in the query response;

determining not to interrupt the user when a private moment is detected; and

determining not to disclose the private information of the user with the virtual assistant.

10. An electronic device for generating a personalized response from a virtual assistant, the electronic device comprising:

a network communication circuitry;

a memory; and

at least one processor configured to:

receive a user query including a wake-word;

replacing unstructured inputs included in the user query with a machine-comprehendible structured form;

separating the wake-word and a remaining portion of the user query;

sending the wake-word to a wake-word processor;

parsing the remaining portion of the user query separately from the wake-word;

retrieve a query response from at least one knowledge base based on the wake-word;

generate a standard natural language response from the query response; and

synthesize a personalized response from a standard natural language response based on the wake-word,

wherein retrieving the query response further comprises:

receiving a plurality of action steps from an action planner module;

searching for an appropriate knowledge base from the user query;

extracting the query response to the user query from the searched appropriate knowledge base;

instructing the query response to be sent to a response path planner module;

searching the user query in a plurality of knowledge bases;

preparing a routing plan for searching the query response in one of the plurality of knowledge bases; and

routing the user query to the appropriate knowledge base for retrieving the query response.

11. The electronic device of claim 10 , wherein the at least one processor is further configured to:

parse the user query to separate the wake-word from the user query;

process the wake-word;

extract wake-word related information from a wake-word database;

parse the extracted wake-word related information along with a plurality of user preference information;

classify information received from a wake-word parser; and

produce a wake-word context from the classified information.

12. The electronic device of claim 11 , wherein the at least one processor is further configured to:

generate a user query context from the wake-word context and separated user query; and

compute the query response based on the user query context.

13. The electronic device of claim 11 , wherein the at least one processor is further configured to:

extract wake-word information from a server through a plurality of application program interface (API) calls; and

receive the wake-word information and user preference information, wherein a data structure of the wake-word information and the user preference information is a standard data format.

14. The electronic device of claim 13 , wherein the server is configured to return the wake-word information and the user preference information to the wake-word parser in response to one of the plurality of API calls.

15. The electronic device of claim 10 , wherein the at least one processor is further configured to:

receive a plurality of wake-words along with a plurality of wake-word parameters, wherein the wake-word parameters are set based on a plurality of user preference information including, at least one of, a response type, a voice of a response, a language, an avatar, or a filler source;

create an entry for the plurality of wake-words along with the plurality of wake-word parameters and a plurality of knowledge base information in a wake-word database, wherein the plurality knowledge base information includes, at least one of, a profile identification (ID), the wake-word, or the knowledge base;

map the plurality of wake-words along with the plurality of knowledge bases; and

store the mapped plurality of wake-words with the plurality of knowledge bases.

16. The electronic device of claim 10 , wherein the at least one processor is further configured to:

interact with a server or a wake-word processor to extract user preference information related to the wake-word;

receive the user preference information;

convert the received user preference information along with a plurality of semantic fillers into a personalized response; and

generate the personalized response to speech in, at least one of, a desired voice, action, form, or text.

17. The electronic device of claim 10 , wherein the at least one processor is further configured to assign a plurality of tags to information received from a wake-word database.

18. The electronic device of claim 10 , wherein a plurality of wake-words are assigned corresponding to at least one mode from among a plurality of modes.

19. The electronic device as claimed in claim 18 , wherein the plurality of modes include a general mode and a private mode, and

wherein the private mode is configured to:

initialize the private mode based on a user query context;

change at least one of a privacy setting, a response setting and a response type;

determine private information of a user and excluding the private information of the user in the query response;

determine not to interrupt the user when a private moment is detected; and

determine not to disclose the private information of the user with the virtual assistant.