Alias resolving intelligent assistant computing device

Patent metadata

Field	Value
Publication number	US-11004446-B2
Application number	US-201715640201-A
Country	US
Kind code	B2
Filing date	Jun 30, 2017
Priority date	Feb 14, 2017
Publication date	May 11, 2021
Grant date	May 11, 2021

Abstract

Intelligent assistant systems, methods and computing devices are disclosed for resolving alias identifiers. A method comprises receiving and parsing data comprising a current user input that includes an alias identifier. The data and/or other sensor data are analyzed to identify the user. Based at least on identifying the user and recognizing the alias identifier, usage pattern data comprising at least one previous user input that includes the alias identifier and corresponding context information is accessed. The usage pattern data is used to resolve the alias identifier to mean the alias identifier in an alias record of a known entity. Based at least on resolving the alias identifier, an output device is controlled to one or more of generate a message and perform an action with respect to the known entity.

First claim

The invention claimed is:

1. An intelligent digital assistant system, comprising:
at least one sensor in an environment;
a logic processor; and
a storage device holding instructions executable by the logic processor to:
via the at least one sensor, receive data comprising a current input provided by a user that includes an alias identifier;
parse the data to recognize the alias identifier;
computer-analyze one or more of the data and other sensor data received from the environment to identify the user as providing the current input;
based at least on identifying the user and recognizing the alias identifier in the data, access usage pattern data of the user comprising at least one previous input provided by the user that includes the alias identifier and corresponding context information;
using at least the previous input and the context information, resolve the alias identifier in the current input to mean the alias identifier in an alias record of a known entity;
based at least on resolving the alias identifier in the current input, control an output device to one or more of generate a message and perform an action with respect to the known entity;
from a new user, receive a new user input that includes the alias identifier;
based at least on usage pattern data of the new user, resolve the alias identifier in the new user input to mean the alias identifier in the alias record of the known entity; and
based at least on resolving the alias identifier in the new user input, classify the new user into a category that relates to one or more aspects of the alias identifier.

2. The intelligent digital assistant system of claim 1 , wherein the at least one sensor comprises a microphone, the data comprises audio data and the current input comprises a spoken utterance of the user.

3. The intelligent digital assistant system of claim 2 , wherein the at least one sensor further comprises a camera, and the data further comprises image data captured with the spoken utterance of the user.

4. The intelligent digital assistant system of claim 1 , wherein the at least one sensor comprises a camera, the data comprises image data and the current input comprises a gesture of the user.

5. The intelligent digital assistant system of claim 1 , wherein the context information is captured contemporaneously with the previous user input provided by the user.

6. The intelligent digital assistant system of claim 1 , wherein the context information comprises one or more of audio data and image data.

7. The intelligent digital assistant system of claim 1 , wherein the instructions are executable to:
from another user receive alias record input comprising the alias identifier; and
populate the alias record with the alias record input from the other user.

8. The intelligent digital assistant system of claim 7 , wherein the alias record input comprises an utterance spoken by the other user.

9. The intelligent digital assistant system of claim 1 , wherein the instructions are executable to:
identify a link between the alias identifier and other information of the user that relates to the one or more aspects of the alias identifier; and
use the link to classify the new user into the category that relates to the one or more aspects of the alias identifier.

10. At a computing device, a method for resolving an alias identifier, the method comprising:
via at least one sensor in an environment, receiving data comprising a current input provided by a user that includes the alias identifier;
parsing the data to recognize the alias identifier;
computer-analyzing one or more of the data and other sensor data received from the environment to identify the user as providing the current input;
based at least on identifying the user and recognizing the alias identifier in the data, accessing usage pattern data of the user comprising at least one previous input provided by the user that includes the alias identifier and corresponding context information;
using at least the previous input and the context information, resolving the alias identifier in the current input to mean the alias identifier in an alias record of a known entity;
based at least on resolving the alias identifier in the current input, controlling an output device to one or more of generate a message and perform an action with respect to the known entity;
from a new user, receiving a new user input that includes the alias identifier;
based at least on usage pattern data of the new user, resolving the alias identifier in the new user input to mean the alias identifier in the alias record of the known entity; and
based at least on resolving the alias identifier in the new user input, classifying the new user into a category that relates to one or more aspects of the alias identifier.

11. The method of claim 10 , wherein the at least one sensor comprises a microphone, the data comprises audio data and the current input comprises a spoken utterance of the user.

12. The method of claim 11 , wherein the at least one sensor further comprises a camera, and the data further comprises image data captured with the spoken utterance of the user.

13. The method of claim 10 , wherein the at least one sensor comprises a camera, the data comprises image data and the current input comprises a gesture of the user.

14. The method of claim 10 , further comprising capturing the context information contemporaneously with the previous user input provided by the user.

15. The method of claim 10 , wherein the context information comprises one or more of audio data and image data.

16. The method of claim 10 , further comprising:
receiving from another user alias record input comprising the alias identifier; and
populating the alias record with the alias record input from the other user.

17. The method of claim 16 , wherein the alias record input comprises an utterance spoken by the other user.

18. A computing device, comprising:
a plurality of sensors;
a logic processor; and
a storage device holding instructions executable by the logic processor to:
via the at least one of the sensors, receive data comprising a current input provided by a user that includes an alias identifier;
convert the data into parsable data;
parse the data to recognize the alias identifier;
computer-analyze one or more of the data and other sensor data received from an environment of the user to identify the user as providing the current input;
based at least on identifying the user and recognizing the alias identifier in the data, access usage pattern data of the user comprising at least one previous input provided by the user that includes the alias identifier and corresponding context information;
access an alias record of a known entity;
using at least the previous input and the context information, resolve the alias identifier in the current input to mean the alias identifier in the alias record of the known entity;
based at least on resolving the alias identifier in the current input, control an output device to one or more of generate a message and perform an action with respect to the known entity;
from a new user, receive a new user input that includes the alias identifier;
based at least on usage pattern data of the new user, resolve the alias identifier in the new user input to mean the alias identifier in the alias record of the known entity; and
based at least on resolving the alias identifier in the new user input, classify the new user into a category that relates to one or more aspects of the alias identifier.

Assignees

• Microsoft Technology Licensing Llc

Inventors

• Finkelstein Erich-Soren
• Fung Han Yee Mimi
• Solomon Oz

Classifications

• G06F18/251

  of input or preprocessed data · CPC title

• G06F18/41

  Interactive pattern learning with a human teacher · CPC title

• G06F18/29

  Graphical models, e.g. Bayesian networks · CPC title

• H04N23/611

  where the recognised objects include parts of the human body · CPC title

• G06F18/28

  Determining representative reference patterns, e.g. by averaging or distorting; Generating dictionaries · CPC title