System and method for correcting accent induced speech in an aircraft cockpit utilizing a dynamic speech database

• US 9,299,340 B2
• Filed: 10/07/2013
• Issued: 03/29/2016
• Est. Priority Date: 10/07/2013
• Status: Active Grant

• Current Assignee: Honeywell International Inc.
• Original Assignee: Honeywell International Inc.
• Inventors: Kar, Satyanarayan, Krishna, Kiran Gopala, Agarwal, Jitender Kumar, De Mers, Robert E.
• Primary Examiner(s): Han, Qi
• Application Number:
  US14/047,518
• Publication Number:
  US 20150100311A1
• Time in Patent Office: 904 Days
• Field of Search: 704/231, 704/235, 704/251, 704/236, 704/246, 704/250, 704/255, 704/270, 704/270.1
• US Class Current: 1/1
• CPC Class Codes: G08G 5/0013 with a ground station
• CPC Class Codes: G08G 5/0021 located in the aircraft
• CPC Class Codes: G10L 15/07 to the speaker
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue

Abstract

A system and method for recognizing speech on board an aircraft that compensates for different regional dialects over an area comprised of at least first and second distinct geographical regions, comprises analyzing speech in the first distinct geographical region using speech data characteristics representative of speech in the first distinct geographical region, detecting a change in position from the first distinct geographical region to the second geographical region, and analyzing speech in the second distinct geographical region using speech data characteristics representative of speech in the second distinct geographical region upon detecting that the aircraft has transitioned from the first distinct geographical region to the second distinct geographical region.

Claims

1. A method, executable by a processor, for recognizing speech in a speech recognition system on board an aircraft that compensates for different regional dialects over an area comprised of at least first and second distinct geographical regions, the method comprising:

analyzing speech in the first distinct geographical region using speech data characteristics representative of speech in the first distinct geographical region, wherein analyzing the speech data characteristics comprises referencing at least one of an acoustic model, a phonetic dictionary, and a language model in the speech recognition system on board of the aircraft;

detecting a change in position from the first distinct geographical region to the second geographical region; and

analyzing speech in the second distinct geographical region using speech data characteristics representative of speech in the second distinct geographical region upon detecting that the aircraft has transitioned from the first distinct geographical region to the second distinct geographical region.

2. The method of claim 1 wherein the first distinct geographic region corresponds to an initial position of the aircraft.

3. The method of claim 1 further comprising:

listening to air traffic control samples;

constructing a speech database for each distinct geographic region; and

selecting a speech database that provides the greatest accuracy.

4. The method of claim 3 wherein the speech database is selected for a flight information region (FIR) corresponding to one of an active communication frequency and a standby communication frequency.

5. The method of claim 4 further comprising loading a new speech database when the aircraft enters a new distinct geographic region.

6. A method, executable by a processor, for recognizing speech in a speech recognition system on board an aircraft that compensates for different regional dialects over an area comprised of at least first and second distinct geographical regions, the method comprising:

analyzing speech in the first distinct geographical region using speech data characteristics representative of speech in the first distinct geographical region;

detecting a change in position from the first distinct geographical region to the second geographical region; and

analyzing speech in the second distinct geographical region using speech data

characteristics representative of speech in the second distinct geographical region upon detecting that the aircraft has transitioned from the first distinct geographical region to the second distinct geographical region;

wherein each distinct geographical region is characterized by a unique aircraft communication frequency, and further comprising;

monitoring the aircraft communication frequency;

detecting a new communication frequency when the aircraft transitions from the first distinct geographical region to the second distinct geographical region; and

analyzing speech in the second distinct geographical region using the speech data characteristics representative of speech in the second distinct geographical region.

7. A method, executable by a processor, for recognizing speech in a speech recognition system on board an aircraft that compensates for different regional dialects over an area comprised of at least first and second distinct geographical regions, the method comprising:

analyzing speech in the first distinct geographical region using speech data characteristics representative of speech in the first distinct geographical region;

detecting a change in position from the first distinct geographical region to the second geographical region; and

analyzing speech in the second distinct geographical region using speech data

characteristics representative of speech in the second distinct geographical region upon detecting that the aircraft has transitioned from the first distinct geographical region to the second distinct geographical region;

wherein each distinct geographical region is characterized by a unique aircraft communication frequency, and further comprising;

monitoring the aircraft communication frequency, wherein the aircraft communication frequency is one of an active frequency and a standby frequency;

detecting a new communication frequency when the aircraft transitions from the first distinct geographical region to the second distinct geographical region; and

analyzing speech in the second distinct geographical region using the speech data characteristics representative of speech in the second distinct geographical region.

8. A method, executable by a processor, for recognizing speech in a speech recognition system on board an aircraft that compensates for different regional dialects over an area comprised of at least first and second distinct geographical regions, the method comprising:

analyzing speech in the first distinct geographical region using speech data characteristics representative of speech in the first distinct geographical region;

detecting a change in position from the first distinct geographical region to the second geographical region; and

analyzing speech in the second distinct geographical region using speech data

characteristics representative of speech in the second distinct geographical region upon detecting that the aircraft has transitioned from the first distinct geographical region to the second distinct geographical region;

wherein each distinct geographical region is characterized by a unique aircraft communication frequency, and further comprising;

monitoring the aircraft communication frequency, wherein the aircraft communication frequency is one of an active frequency and a standby frequency;

monitoring a voice feed from air traffic control;

detecting a new communication frequency when the aircraft transitions from the first distinct geographical region to the second distinct geographical region; and

analyzing speech in the second distinct geographical region using the speech data characteristics representative of speech in the second distinct geographical region.

9. A method, executable by a processor, for recognizing speech in a speech recognition system on board an aircraft that compensates for different regional dialects over an area comprised of at least first and second distinct geographical regions, the method comprising:

analyzing speech in the first distinct geographical region using speech data characteristics representative of speech in the first distinct geographical region;

detecting a change in position from the first distinct geographical region to the second geographical region; and

analyzing speech in the second distinct geographical region using speech data

characteristics representative of speech in the second distinct geographical region upon detecting that the aircraft has transitioned from the first distinct geographical region to the second distinct geographical region;

wherein each distinct geographical region is characterized by a unique aircraft communication frequency, and further comprising;

monitoring the aircraft communication frequency, wherein the aircraft communication frequency is one of an active frequency and a standby frequency;

monitoring a voice feed from air traffic control;

detecting a new communication frequency when the aircraft transitions from the first distinct geographical region to the second distinct geographical region;

analyzing speech in the second distinct geographical region using the speech data characteristics representative of speech in the second distinct geographical region; and

displaying air traffic control text on a cockpit display.

10. The method of claim 9 wherein a selected speech database for a region is the one with the greatest accuracy for that region.

11. The method of claim 10 further comprising initializing the speech recognition system with initial position data and an initial flight information region (FIR).

12. An aircraft speech recognition system that compensates for regional dialects over an area comprised of at least first and second distinct geographical regions, comprising:

a first source of speech data that represents a first acoustic model, a first phonetic dictionary, and a first language model in the aircraft speech recognition system that is characteristic of speech in the first distinct geographical region;

a second source of speech data that represents a second acoustic model, a second phonetic dictionary, and a second language model in the aircraft speech recognition system that is characteristic of speech in the second distinct geographical region; and

a processor coupled to the first source and to the second source and configured to process speech in the first distinct geographical region utilizing data in the first source, switch to the second source when the aircraft enters the second distinct geographical region, and process speech in the second distinct geographical region utilizing data in the second source.

13. The system of claim 12 wherein the first data source comprises a speech database and a flight information region (FIR) database associated with each geographic region, an wherein the speech database for each geographic region comprises an acoustic model optimized for each FIR region.

14. The system of claim 13 wherein the processor is coupled to the FIR database, the speech database, a third source of aircraft position data, and a fourth source of aircraft communication frequency data.

15. The system of claim 14 wherein the fourth source comprises active and standby frequency data.

16. The system of claim 15 further comprising a cockpit display device coupled to the processor for displaying processed speech.

17. The system of claim 16 further comprising an air traffic control audio input coupled to the processor.

18. The system of claim 13 wherein the FIR database contains the geographic boundaries of each FIR region.