Systems and methods for aviation communication monitoring using speech recognition

Patent metadata

Field	Value
Publication number	US-9466290-B2
Application number	US-201414270519-A
Country	US
Kind code	B2
Filing date	May 6, 2014
Priority date	May 6, 2014
Publication date	Oct 11, 2016
Grant date	Oct 11, 2016

Abstract

Systems and methods for aviation communication monitoring using speech recognition are provided. In one embodiment, a method for multichannel avionics communication is provided. The method comprises: configuring a software defined radio to generate digitized audio data for a plurality of voice communication channels; audibly reproducing voice communication received over a first channel of the plurality of voice communication channels; applying a trigger word speech recognition algorithm to at least a second channel of the plurality of voice communication channels to detect a message containing a trigger word; and when a trigger word is detected, storing a communication containing the trigger word and generating an alert.

First claim

What is claimed is:

1. A multichannel avionics communication system, the system comprising:
a software defined radio coupled to an antenna, where the software defined radio produces digitized audio data for a plurality of voice communication channels based on digital samples of radio frequency (RF) carrier signals received by the antenna;
an audio panel device coupled to the software defined radio and at least one speaker, the audio panel device receiving the digitized audio data for the plurality of voice communication channels, wherein using the at least one speaker, the audio panel device audibly reproduces, from a first portion of the digitized audio data, voice communication received over a first channel of the plurality of voice communication channels, the first channel corresponding to a user selected tuned frequency; and
a processor executing a trigger word speech recognition algorithm that receives the digitized audio data and processes at least a second channel of the plurality of voice communication channels from a second portion of the digitized audio data to detect a message containing a trigger word, the at least a second channel corresponding to a monitored frequency;
wherein when the trigger word speech recognition algorithm detects a trigger word, a communication associated with the trigger word and received on the at least a second channel is stored into a memory and an alert is generated.

2. The system of claim 1 , wherein either the software defined radio or the audio panel device comprises the processor executing the trigger word speech recognition algorithm.

3. The system of claim 1 , wherein the alert comprises an audible alert.

4. The system of claim 1 , wherein the alert comprises a message displayed on a display device.

5. The system of claim 4 , wherein the message displayed on the display device includes an indication identifying the trigger word.

6. The system of claim 4 , wherein the display device comprises an avionics multi-function display (MFD) unit or an avionics Navigator console.

7. The system of claim 1 , further comprising a human-machine interface device providing a selection option for a user to recall and listen to the stored communication associated with the trigger word.

8. The system of claim 7 , the memory including a buffer storing the digitized audio data so that the user can listen to at least a portion of the communication that occurred prior to occurrence of the trigger word.

9. The system of claim 1 , further comprising:
a flight management system coupled to the software defined radio, wherein the flight management system configures the software defined radio as to which RF carrier signals corresponding are included in a set of monitored frequencies.

10. The system of claim 9 , wherein the flight management system automatically selects the set of monitored frequencies to include at least one of:
an emergency channel;
a prior channel that was a previous user selected tuned frequency;
a channel that is a next anticipated control center handoff; or
a Universal Communications (Unicom) stations.

11. The system of claim 9 , wherein the flight management system automatically selects the set of monitored frequencies based on at least one of:
an aircraft position; or
a flight stage for the aircraft.

12. A method for multichannel avionics communication, the method comprising:
configuring a software defined radio to generate digitized audio data for a plurality of voice communication channels;
audibly reproducing voice communication received over a first channel of the plurality of voice communication channels from a first portion the digitized audio data;
applying a trigger word speech recognition algorithm to at least a second channel of the plurality of voice communication channels from a second portion of the digitized audio data to detect a message containing a trigger word; and
when a trigger word is detected, storing a communication containing the trigger word and generating an alert.

13. The method of claim 12 , further comprising:
audibly reproducing the communication when a user selects the communication for playback.

14. The method of claim 12 , wherein the trigger word speech recognition algorithm is implemented either:
by a processor of the software defined radio; or
by a processor of an audio panel.

15. The method of claim 12 , wherein generating an alert comprises one or both of:
generating an audible alert; and
displaying a message on a display device.

16. The method of claim 15 , wherein the message displayed on the display device includes an indication identifying the trigger word.

17. The method of claim 15 , wherein displaying the message on the display device further comprises:
providing a selection option for a user to recall and listen to the stored communication associated with the trigger word.

18. The method of claim 12 , wherein configuring the software defined radio to generate digitized audio data for a plurality of voice communication channels further comprises:
identifying to the software defined radio a set of monitored frequencies that include at least one of:
an emergency channel;
a prior channel that was a previous user selected tuned frequency;
a channel that is a next anticipated control center handoff; or
a Universal Communications (Unicom) stations.

19. The method of claim 12 , wherein configuring the software defined radio to generate digitized audio data for a plurality of voice communication channels is based on one of both of:
an aircraft position; or
a flight stage for the aircraft.

20. The method of claim 12 , further comprising:
buffering the digitized audio data so that a user can listen to at least a portion of the communication that occurred prior to occurrence of the trigger word.

Assignees

• Honeywell Int Inc

Inventors

• Joyce Jim

Classifications

• G10L2015/088

  Word spotting · CPC title

• G10L15/22

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

• G10L15/08 Primary

  Speech classification or search · CPC title