Adaptively learning vocabulary for completing speech recognition commands

Patent metadata

Field	Value
Publication number	US-10770060-B2
Application number	US-201314097878-A
Country	US
Kind code	B2
Filing date	Dec 5, 2013
Priority date	Dec 5, 2013
Publication date	Sep 8, 2020
Grant date	Sep 8, 2020

Abstract

An embodiment provides a method, including: receiving, via an audio receiver of an information handling device, user voice input; identifying a first word based on the user voice input; accessing a word association data store; selecting an equivalent based on an association with the first word within the word association data store; committing an action based on the equivalent; receiving feedback input from the user regarding the equivalent; and updating the selecting based on the feedback. Other aspects are described and claimed.

First claim

What is claimed is:

1. A method, comprising:
receiving, via an audio receiver of an information handling device, user voice input from a user;
accessing, using a processor, a user profile associated with the user;
identifying, using a processor, a first word from the user voice input, wherein the first word demands performance of an original action or a user-specific action, wherein the original action performs a different command than the user-specific action;
accessing, using the processor, a word association data store associated with the user profile;
determining, in the word association data store, whether feedback input adjusting the original action to the user-specific action for the first word was previously received, wherein the feedback input does not adjust an identification of the first word; and
performing, responsive to determining that the feedback input was previously received, the user-specific action.

2. The method of claim 1 , wherein the feedback input corresponds to negative feedback input.

3. The method of claim 1 , wherein the word associations within the word association data store are word associations selected from the group of associations consisting of colloquialisms associated with formal language equivalents, slang terms associated with formal language terms, jargon terms associated with formal language terms, and dialect specific terms associated with formal language terms.

4. The method of claim 1 , wherein the word association data store includes formal language equivalents for a plurality of formal languages.

5. The method of claim 1 , further comprising providing a communication to the user prior to executing the user-specific action.

6. The method of claim 5 , wherein the communication queries the user for confirmation.

7. The method of claim 1 , wherein the feedback input was previously received from a group of associated users.

8. The method of claim 7 , wherein the group of associated users comprises users associated via a cloud account linkage.

9. An information handling device, comprising:
an audio receiver;
a processor; and
a memory device that stores instructions executable by the processor to:
receive, via the audio receiver, user voice input from a user;
access a user profile associated with the user;
identify, using the processor, a first word from the user voice input, wherein the first word demands performance of an original action or a user-specific action, wherein the original action performs a different command than the user-specific action;
access, using the processor, a word association data store associated with the user profile;
determine, in the word association data store, whether feedback input adjusting the original action to the user-specific action for the first word was previously received, wherein the feedback input does not adjust an identification of the first word; and
perform, responsive to determining that the feedback input was previously received, the user-specific action.

10. The information handling device of claim 9 , wherein the feedback input corresponds to negative feedback input.

11. The information handling device of claim 9 , wherein the word associations within the word association data store are word associations selected from the group of associations consisting of colloquialisms associated with formal language equivalents, slang terms associated with formal language terms, jargon terms associated with formal language terms, and dialect specific terms associated with formal language terms.

12. The information handling device of claim 9 , wherein the word association data store includes formal language equivalents for a plurality of formal languages.

13. The information handling device of claim 9 , wherein the instructions are further executable by the processor to provide a communication to the user prior to executing the user-specific action.

14. The information handling device of claim 13 , wherein the communication queries the user for confirmation.

15. The information handling device of claim 13 , wherein the communication comprises a list of suggested commands to be executed.

16. The information handling device of claim 1 , wherein the feedback input was previously received from a group of associated users.

17. A product, comprising:
a storage device having code stored therewith, the code comprising:
code that receives, via an audio receiver of an information handling device, user voice input from a user;
code that access a user profile associated with the user;
code that identifies a first word from the user voice input, wherein the first word demands performance of an original action or a user-specific action, wherein the original action performs a different command than the user-specific action;
code that accesses a word association data store associated with the user profile;
code that determines, in the word association data store, whether feedback input adjusting the original action to the user-specific action for the first word was previously received, wherein the feedback input does not adjust an identification of the first word; and
code that performs, responsive to determining that the feedback input was previously received, the user-specific action.

Assignees

• Lenovo Singapore Pte Ltd

Inventors

• Vanblon Russell Speight
• Heim Jon Wayne
• Knox Jonathan Gaither
• Wetsel Peter Hamilton
• Beaumont Suzanne Marion

Classifications

• G10L15/063 Primary

  Training · CPC title

• G10L15/22

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

• G10L2015/223

  Execution procedure of a spoken command · CPC title

• G10L2015/0638

  Interactive procedures · CPC title