Curiosity based activation and search depth

Patent metadata

Field	Value
Publication number	US-11769501-B2
Application number	US-202117303552-A
Country	US
Kind code	B2
Filing date	Jun 2, 2021
Priority date	Jun 2, 2021
Publication date	Sep 26, 2023
Grant date	Sep 26, 2023

Abstract

Embodiments of the present invention determine a curiosity of a user based on data received from an electronic device associated with the user, where the data includes audible speech captured from user and one or more facial expressions of the user. Embodiments of the present invention identify a first wavelength for audible speech from the user to initiate a command detection mode based on a plurality of wavelengths associated with a user profile for the user. Embodiments of the present invention identify a topic for the audible speech from the user and responsive to determining an intelligent virtual assistant is an intended recipient based on the topic, suspend an activation word for the intelligent virtual assistant.

First claim

What is claimed is:

1. A computer-implemented method comprising:
determining a curiosity of a user based on data received from an electronic device associated with the user, wherein the data includes audible speech captured from the user and one or more facial expressions of the user;
determining a grade of curiosity based on the data received from the electronic device associated with the user;
identifying a first wavelength for the audible speech from the user to initiate a command detection mode based on a plurality of wavelengths associated with a user profile for the user;
identifying a topic for the audible speech from the user;
responsive to determining an intelligent virtual assistant is an intended recipient based on the topic, suspending an activation word for the intelligent virtual assistant;
determining a search depth score for a command performable by the intelligent virtual assistant within the audible speech of the user; and
responsive to sending the search depth score for the command performable by the intelligent virtual assistant to a backend search engine, providing a response time for executing the command performable by the intelligent virtual assistant to the user based on the search depth score.

2. The computer-implemented method of claim 1 , further comprising:
identifying a first set of wavelengths from the plurality of wavelengths associated with the user profile for the user, wherein the first set of wavelengths are associated with a first plurality of questions presented to the intelligent virtual assistant;
performing a wavelength clarification test to identify a second set of wavelengths from the plurality of wavelengths associated with the user, wherein the second set of wavelengths are associated with a second plurality of questions not presented to the intelligent virtual assistant; and
establishing the user profile for the user based on the first set of wavelengths and the second set of wavelengths.

3. The computer-implemented method of claim 2 , wherein performing the wavelength clarification test further comprises:
sending a plurality of sample questions to the intelligent virtual assistant;
instructing the intelligent virtual assistant to request the user audibly repeat the plurality of sample questions, wherein the plurality of sample questions do not relate to a command performable by the intelligent virtual assistant; and
identifying the second set of wavelengths for the plurality of sample questions audibly repeated by the user.

4. The computer-implemented method of claim 1 , further comprising:
receiving a modification to the search depth score from the user, wherein a modified search depth score alters the response time for executing the command performable by the intelligent virtual assistant.

5. The computer-implemented method of claim 1 , further comprising:
analyzing a command performable by the intelligent virtual assistant within the audible speech of the user;
determining at least two knowledge bases for the command performable by the intelligent virtual assistant, wherein a backend search engine utilizes the at least two knowledge bases to search for a response to the command performable by the intelligent virtual assistant; and
responsive to determining parallelization for the at least two knowledge bases are supported, parallelizing the search with the at least two knowledge bases.

6. The computer-implemented method of claim 1 , further comprising:
analyzing a command performable by the intelligent virtual assistant within the audible speech of the user;
determining at least two knowledge bases for the command performable by the intelligent virtual assistant, wherein a backend search engine utilizes the at least two knowledge bases to search for a response to the command performable by the intelligent virtual assistant;
responsive to determining parallelization for the at least two knowledge bases is not supported, loading a first knowledge base into the backend search engine; and
sending a result from the first knowledge base to a second knowledge base.

7. A computer program product comprising:
one or more computer readable storage media and program instructions stored on the one or more computer readable storage media, the program instructions comprising:
program instructions to determine a curiosity of a user based on data received from an electronic device associated with the user, wherein the data includes audible speech captured from the user and one or more facial expressions of the user;
program instructions to determine a grade of curiosity based on the data received from the electronic device associated with the user;
program instructions to identify a first wavelength for the audible speech from the user to initiate a command detection mode based on a plurality of wavelengths associated with a user profile for the user;
program instructions to identify a topic for the audible speech from the user;
program instructions to, responsive to determining an intelligent virtual assistant is an intended recipient based on the topic, suspend an activation word for the intelligent virtual assistant;
program instructions to determine a search depth score for a command performable by the intelligent virtual assistant within the audible speech of the user; and
program instructions to, responsive to sending the search depth score for the command performable by the intelligent virtual assistant to a backend search engine, provide a response time for executing the command performable by the intelligent virtual assistant to the user based on the search depth score.

8. The computer program product of claim 7 , wherein the program instructions stored on the one or more computer readable storage media further comprise:
program instructions to identify a first set of wavelengths from the plurality of wavelengths associated with the user profile for the user, wherein the first set of wavelengths are associated with a first plurality of questions presented to the intelligent virtual assistant;
program instructions to perform a wavelength clarification test to identify a second set of wavelengths from the plurality of wavelengths associated with the user, wherein the second set of wavelengths are associated with a second plurality of questions not presented to the intelligent virtual assistant; and
program instructions to establish the user profile for the user based on the first set of wavelengths and the second set of wavelengths.

9. The computer program product of claim 8 , wherein the program instructions to perform the wavelength clarification test comprises:
program instructions to send a plurality of sample questions to the intelligent virtual assistant;
program instructions to instruct the intelligent virtual assistant to request the user audibly repeat the plurality of sample questions, wherein the plurality of sample questions do not relate to a command performable by the intelligent virtual assistant; and
program instructions to identify the second set of wavelengths for the plurality of sample questions audibly repeated by the user.

10. The computer program product of claim 7 , wherein the program instructions stored on the one or more computer readable storage media further comprise:
program instructions to receive a modification to the search depth score from the user, wherein a modified search depth score alters the response time for executing the command performable by the intelligent virtual assistant.

11. The computer program product of claim 7 , wherein the program instructions stored on the one or more computer readable storage media further comprise:
program instructions to analyze a comm

Assignees

• IBM

Inventors

• Eda Sasikanth
• Rakshit Sarbajit K
• Jain Abhishek
• Patil Sandeep Ramesh

Classifications

• G10L15/22 Primary

  Procedures used during a speech recognition process, e.g. man-machine dialogue · CPC title

• G10L15/18

  using natural language modelling · CPC title

• G10L15/25

  using position of the lips, movement of the lips or face analysis · CPC title

• G10L25/63 Primary

  for estimating an emotional state · CPC title