SYSTEM AND METHOD FOR CORRECTING SPEECH INPUT

• US 20160004502A1
• Filed: 09/15/2015
• Published: 01/07/2016
• Est. Priority Date: 07/16/2013
• Status: Abandoned Application

• Current Assignee: Lenny Insurance Limited (TATA Motors Limited)
• Original Assignee: CloudCar, Inc.
• Inventors: Othmer, Konstantin, Winkelman, Dominic, Eide, Daniel
• Application Number:
  US14/855,295
• Publication Number:
  US 20160004502A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 3/167 Audio in a user interface, e.g. using voice commands for navigating, audio feedback
• CPC Class Codes: G10L 15/187 Phonemic context, e.g. pronunciation rules, phonotactical constraints or phoneme n-grams
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue

Abstract

A system and method for correcting speech input are disclosed. A particular embodiment includes: receiving a base input string; detecting a correction operation; receiving a replacement string in response to the correction operation; generating a base object set from the base input string and a replacement object set from the replacement string; identifying a matching base object of the base object set that is most phonetically similar to a replacement object of the replacement object set; and replacing the matching base object with the replacement object in the base input string.

Claims

1. A system comprising:

a data processor; and

a speech input processing module, executable by the data processor, the speech input processing module being configured to;

receive a base input string;

detect a correction operation;

receive a replacement string in response to the correction operation;

generate a base object set from the base input string and a replacement object set from the replacement string;

identify a matching base object of the base object set that is most phonetically similar to a replacement object of the replacement object set; and

replace the matching base object with the replacement object in the base input string.

2. The system of claim 1 wherein the base input string is received as a spoken utterance.

3. The system of claim 1 wherein the correction operation is explicitly initiated by use of an input mechanism from the group consisting of:

clicking an icon, activating a softkey, pressing a physical button, providing a keyboard input, manipulating a user interface, and uttering a separate audible command.

4. The system of claim 1 wherein the correction operation is implicitly initiated by detection of a speaker audibly spelling out a word or phrase.

5. The system of claim 1 wherein the replacement is received as a spoken utterance.

6. The system of claim 1 being further configured to generate a phonetic representation of each of a plurality of objects in the base object set.

7. The system of claim 1 being further configured to generate a phonetic representation of each of a plurality of objects in the replacement object set.

8. The system of claim 1 being further configured to generate a difference score between each of a plurality of objects in the base object set and each of a plurality of objects in the replacement object set.

9. The system of claim 1 wherein the speech input processing module is included in an application (app) executed on a platform from the group consisting of:

a mobile device, an in-vehicle control system, and a network service in a network cloud.

10. A method comprising:

receiving a base input string;

detecting a correction operation;

receiving a replacement string in response to the correction operation;

generating a base object set from the base input string and a replacement object set from the replacement string;

identifying a matching base object of the base object set that is most phonetically similar to a replacement object of the replacement object set; and

replacing the matching base object with the replacement object in the base input string.

11. The method of claim 10 wherein the base input string is received as a spoken utterance.

12. The method of claim 10 wherein the correction operation is explicitly initiated by use of an input mechanism from the group consisting of:

clicking an icon, activating a softkey, pressing a physical button, providing a keyboard input, manipulating a user interface, and uttering a separate audible command.

13. The method of claim 10 wherein the correction operation is implicitly initiated by detection of a speaker audibly spelling out a word or phrase.

14. The method of claim 10 wherein the replacement string is received as a spoken utterance.

15. The method of claim 10 including generating a phonetic representation of each of a plurality of objects in the base object set.

16. The method of claim 10 including generating a phonetic representation of each of a plurality of objects in the replacement object set.

17. The method of claim 10 including generating a difference score between each of a plurality of objects in the base object set and each of a plurality of objects in the replacement object set.

18. The method of claim 10 wherein the method is performed by an application (app) executed on a platform from the group consisting of:

a mobile device, an in-vehicle control system, and a network service in a network cloud.

19. A non-transitory machine-useable storage medium embodying instructions which, when executed by a machine, cause the machine to:

receive a base input string;

detect a correction operation;

receive a replacement string in response to the correction operation;

generate a base object set from the base input string and a replacement object set from the replacement string;

identify a matching base object of the base object set that is most phonetically similar to a replacement object of the replacement object set; and

replace the matching base object with the replacement object in the base input string.

20. The machine-useable storage medium as claimed in claim 19 wherein the instructions are included in an application (app) executed on a platform from the group consisting of:

a mobile device, an in-vehicle control system, and a network service in a network cloud.