Methods and systems for utilizing voice commands onboard an aircraft

• US 8,515,763 B2
• Filed: 11/24/2009
• Issued: 08/20/2013
• Est. Priority Date: 11/24/2009
• Status: Active Grant

• Current Assignee: United States Of America As Represented By The Secretary Of The Army (Government of the United States of America), Honeywell International Inc.
• Original Assignee: Honeywell International Inc.
• Inventors: Dong, Xian Qin, Qin, Xiao Long
• Primary Examiner(s): AZAD, ABUL K
• Application Number:
  US12/625,186
• Publication Number:
  US 20110125503A1
• Time in Patent Office: 1,365 Days
• Field of Search: 704/226, 704/233, 704/275, 701/14
• US Class Current: 704/275
• CPC Class Codes: G08G 5/0021 located in the aircraft
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command
• CPC Class Codes: G10L 2015/225 Feedback of the input speech
• CPC Class Codes: G10L 2015/228 of application context
• CPC Class Codes: G10L 21/0208 Noise filtering
• CPC Class Codes: G10L 21/0216 characterised by the method used for estimating noise
• CPC Class Codes: G10L 21/06 Transformation of speech into a non-audible representation, e.g. speech visualisation or speech processing for tactile aids G10L15/26 takes precedence

Abstract

Methods and systems are provided for utilizing audio commands onboard an aircraft. A method comprises identifying a flight phase for the aircraft, resulting in an identified flight phase, receiving an audio input, resulting in received audio input, filtering the received audio input in a manner that is influenced by the identified flight phase for the aircraft, resulting in filtered audio input, and validating the filtered audio input as a first voice command of a first plurality of possible voice commands.

Claims

1. A method for utilizing audio commands onboard an aircraft, the method comprising:

identifying a flight phase for the aircraft, resulting in an identified flight phase;

receiving an audio input, resulting in received audio input;

filtering the received audio input in a manner that is influenced by the identified flight phase for the aircraft, resulting in filtered audio input; and

validating the filtered audio input as a first voice command of a first plurality of possible voice commands, wherein validating the filtered audio input comprises;

determining a plurality of possible initial command words based on the first plurality of possible voice commands, each possible initial command word of the plurality of possible initial command words corresponding to an initial command word of a respective voice command of the first plurality of possible voice commands;

recognizing a first portion of the filtered audio input as a first command word of the first voice command based on the plurality of possible initial command words;

determining a plurality of possible second command words that may follow the first command word of the first voice command based on a subset of the first plurality of possible voice commands, wherein the first command word corresponds to an initial command word for each possible voice command of a subset of the first plurality of possible voice commands and each possible second command word of the plurality of possible second command words corresponds to a second command word of a respective voice command of the subset; and

recognizing a second portion of the filtered audio input as a second command word of the first voice command based on the plurality of possible second command words.

2. The method of claim 1, wherein filtering the received audio input further comprises filtering the received audio input based on an acoustic model corresponding to the identified flight phase.

3. The method of claim 2, further comprising selecting, from a plurality of possible noise models, the acoustic noise model corresponding to the identified flight phase.

4. The method of claim 2, further comprising augmenting the acoustic noise model with synthesized audio corresponding to an auditory warning from an avionics system onboard the aircraft.

5. The method of claim 2, wherein:

receiving the audio input comprises receiving the audio input using an audio input device onboard the aircraft; and

the acoustic model corresponding to the identified flight phase is based on audible noise proximate the audio input device during the identified flight phase.

6. The method of claim 1, wherein validating the filtered audio input further comprises:

validating the first voice command when the identified flight phase corresponds to an allowed flight phase of one or more allowed flight phases for the first voice command; and

invalidating the first voice command if a flight phase criterion associated with the first voice command does not allow execution during the identified flight phase.

7. The method of claim 1, the first voice command being associated with a criterion corresponding to a range of acceptable values for a parameter of the first voice command, wherein validating the filtered audio input further comprises validating the first voice command when a portion of the filtered audio input corresponding to the parameter is within the range of acceptable values.

8. The method of claim 1, further comprising providing the first voice command to an avionics system onboard the aircraft in response to validating the filtered audio input as the first voice command, wherein the first voice command is executed by the avionics system.

9. The method of claim 1, further comprising arranging the first plurality of possible voice commands in a hierarchical manner, wherein each possible second command word of the plurality of possible second command words has a child-parent relationship with the first command word.

10. The method of claim 9, wherein arranging the first plurality of possible voice commands comprises maintaining the first plurality of possible voice commands in a tree data structure, wherein each possible second command word of the plurality of possible second command words is a child node of the first command word.

11. A method for utilizing audio commands onboard an aircraft, the method comprising:

identifying a flight phase for the aircraft, resulting in an identified flight phase;

receiving an audio input, resulting in received audio input;

filtering the received audio input in a manner that is influenced by the identified flight phase for the aircraft, resulting in filtered audio input;

automatically validating the filtered audio input as a first voice command of a first plurality of possible voice commands when a criterion associated with the first voice command corresponds to a first safety level; and

validating the filtered audio input as the first voice command in response to a confirmation when the criterion associated with the first voice command corresponds to a second safety level.

12. A system for an aircraft comprising:

an avionics system onboard the aircraft;

an audio input device onboard the aircraft, the audio input device being capable of receiving audio input;

a data storage element comprising a first plurality of possible voice commands; and

a processing system coupled to the avionics system, the audio input device, and the data storage element, wherein the processing system is configured to;

identify a flight phase for the aircraft, resulting in an identified flight phase;

filter the audio input based on the identified flight phase, resulting in filtered audio input;

recognize the filtered audio input as a first voice command of the first plurality of possible voice commands by;

determining a plurality of possible initial command words based on the first plurality of possible voice commands;

recognizing a first portion of the filtered audio input as a first command word of a first voice command of the first plurality of possible voice commands based on the plurality of possible initial command words;

determining a plurality of possible second command words that may follow the first command word of the first voice command based on a subset of the first plurality of possible voice commands, wherein the first command word corresponds to an initial command word for each possible voice command of the subset; and

recognizing a second portion of the filtered audio input as a second command word of the first voice command based on the plurality of possible second command words; and

provide the first voice command to the avionics system in response to recognizing the filtered audio input as the first voice command, wherein the avionics system is configured to execute the first voice command.

13. The system of claim 12, wherein:

the data storage element comprises a plurality of noise models, each noise model being based on ambient auditory noise proximate the audio input device during a respective flight phase of the aircraft; and

the processing system is configured to;

select a noise model from the plurality of noise models corresponding to the identified flight phase, resulting in a selected noise model; and

filter the audio input based on the selected noise model.

14. The system of claim 13, further comprising a second avionics system configured to generate an auditory warning comprising synthesized audio output, wherein the processing system is configured to:

augment the selected noise model with the synthesized audio output, resulting in an augmented noise model; and

filter the audio input based on the augmented noise model.

15. The system of claim 12, wherein:

the first plurality of possible voice commands are maintained by the data storage element in a tree data structure; and

each possible second command word of the plurality of possible second command words has a child-parent relationship with the first command word.