Natrual voice speech recognition for flight deck applications

Patent metadata

Field	Value
Publication number	US-9830910-B1
Application number	US-201314038249-A
Country	US
Kind code	B1
Filing date	Sep 26, 2013
Priority date	Sep 26, 2013
Publication date	Nov 28, 2017
Grant date	Nov 28, 2017

Abstract

A computer system receives a voice command and applies one or more contextual filters produce avionics commands based on the voice command. Elements of the voice command are displayed for acceptance by a pilot before being implemented by an avionics system. Individual elements are reevaluated as necessary. Contextual filters include flight paths, flight phase, aircraft location, current weather conditions or information associated with a particular airport.

First claim

What is claimed is:

1. A computer system comprising:
a processor;
memory connected to the processor;
a display connected to the processor;
a receiver connected to the processor;
a microphone connected to the processor; and
computer executable program code configured to execute on the processor,
wherein the computer executable program code is configured to:
continuously monitor air-traffic communications frequencies for an aircraft identification code;
receive air traffic controller data corresponding to a voice command received from an air traffic controller through the receiver, associated with the aircraft identification code;
parse the air traffic controller voice data into a plurality of words corresponding to individual command elements;
correlate individual voice command elements based on air traffic controller data and individual voice command elements based on pilot voice data;
receive pilot voice data corresponding to a pilot voice command from the microphone;
parse the pilot voice data into a plurality of words corresponding to individual command elements;
produce a contextual filter based on a pointer location in a graphical user interface, said pointer location associated with a weighted list of potential voice commands, a location of the aircraft, a flight plan, and at least one weather condition;
produce a dynamic contextual filter based on a flight phase selected from a list including taxiing, airborne, and final approach;
apply the contextual filter and dynamic contextual filter to one or more databases of command elements to find a first word of the plurality of words and define a first command element associated with the first word;
add the first command element to the dynamic contextual filter;
associate at least one command element corresponding to an airport identification with at least one predefined descriptor based on a distance from an airport identified by the processor, and available airport facilities;
produce a plurality of avionics commands based on the individual command elements; and
receive one or more confirmations, each of the one or more confirmations corresponding to an avionics command.

2. The computer system of claim 1 , wherein the computer executable program code is further configured to:
determine that at least one of the plurality of avionics commands does not accurately correspond to an individual command element; and
receive a corrected avionics command.

3. The computer system of claim 1 , wherein displaying the plurality of avionics commands comprises highlighting one or more commands in a predefined color indicating the computer system will take action on the highlighted commands when confirmed.

4. The computer system of claim 1 ,
wherein the computer executable program code is further configured to:
filter a voice command database based on the contextual filter; and
search for at least one word in the voice command database to produce the one or more avionics commands.

5. The computer system of claim 4 ,
wherein the contextual filter is based on the airport property.

6. An aircraft computer system comprising:
a processor;
memory connected to the processor;
an antenna connected to the processor;
a microphone connected to the processor; and
computer executable program code configured to execute on the processor,
wherein the computer executable program code is configured to:
continuously monitor air-traffic communications frequencies for an aircraft identification code;
receive air traffic controller data corresponding to a voice command received from an air traffic controller through the antenna, associated with the aircraft identification code;
parse the air traffic controller voice data into a plurality of words corresponding to individual command elements;
correlate individual voice command elements based on air traffic controller data and individual voice command elements based on pilot voice data;
receive pilot voice data corresponding to a pilot voice command through the microphone;
parse the pilot voice data into a plurality of words corresponding to individual command elements;
associate at least one command element corresponding to an airport identification with at least one predefined descriptor based on a distance from an airport identified by the processor, and available airport facilities;
produce a contextual filter based on a pointer location in a graphical user interface, said pointer location associated with a weighted list of potential voice commands, a location of the aircraft, a flight plan, and at least one weather condition;
produce a dynamic contextual filter based on a flight phase selected from a list including taxiing, airborne, and final approach;
filter a voice command database based on the contextual filter and dynamic contextual filter;
search for at least one word in the voice command database; and
add the at least one word to the dynamic contextual filter.

7. The aircraft computer system of claim 6 , wherein the contextual filter is based on the airport property.

8. The aircraft computer system of claim 6 , wherein the computer executable program code is further configured to:
produce a plurality of avionics commands based on the individual command elements;
display the plurality of avionics commands on the display;
mirror the plurality of avionics commands on a co-pilot display; and
receive one or more confirmations, each of the one or more confirmations corresponding to an avionics command.

9. The aircraft computer system of claim 8 , wherein the computer executable program code is further configured to:
determine that at least one of the plurality of avionics commands does not accurately correspond to an individual command element; and
receive a corrected avionics command.

10. The aircraft computer system of claim 8 , wherein displaying the plurality of avionics commands comprises highlighting one or more commands in a predefined color indicating the computer system will take action on the highlighted commands when confirmed.

11. A method for producing avionics commands through voice recognition comprising:
continuously monitoring air-traffic communications frequencies for an aircraft identification code;
receiving air traffic controller data corresponding to a voice command received from an air traffic controller;
parsing the air traffic controller voice data into a plurality of words corresponding to individual command elements;
correlating individual voice command elements based on air traffic controller data and individual voice command elements based on pilot voice data;
receiving pilot voice data corresponding to a pilot voice command;
parsing the pilot voice data into a plurality of words corresponding to individual command elements;
associating at least one command element corresponding to an airport identification with at least one predefined descriptor based on a distance from an airport identified by the processor, and available airport facilities;
producing a contextual filter based on a pointer location in a graphical user interface, said pointer location associated with a weighted list of potential voice commands, a location of the aircraft, a flight plan, and at least one weather condition;
producing a dynamic contextual filter based on a flight phase selected from a list including taxiing, airborne, and final approach;
filtering a voice command database based on the contextual filter and dynamic contextual filter;
searching for at least one word in the voice command database; and
adding the at least one word to the dynamic contextual filter.

12. The method of claim 11 , wherein the

Assignees

• Rockwell Collins Inc

Inventors

• Shapiro Geoffrey A
• Leedom David L
• Lorch Nicholas M

Classifications

• G10L15/265 Primary

  Physics · mapped topic

• G10L2015/228

  of application context · CPC title

• G10L2015/223

  Execution procedure of a spoken command · CPC title

• G10L15/22 Primary

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

• G08G5/26

  Transmission of traffic-related information between aircraft and ground stations · CPC title