Aircraft Radio Voice Message Post-Processing for Communication Filtering

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current speech recognition systems onboard aircraft struggle to accurately identify and filter out radio communications directed specifically to a particular aircraft from a multitude of communications received, often due to incomplete or misinterpreted identifiers, leading to irrelevant information being presented to flight crews.

Innovation Solution

A method and system that utilize secondary sources, such as remote servers and aircraft transponders, for post-processing speech recognition data to enhance accuracy by comparing interpreted speech data against stored identifiers and secondary source data, and using distinguishing visual characteristics to present relevant communications, even when confidence levels are below a threshold.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If speech recognition systems process all received radio communications onboard an aircraft, then the system can capture all communications transmitted from ground control to multiple aircraft, but the first aircraft receives and processes a significant number of radio communications that are not applicable to its own operations

Engineering Contradiction:
Improvequantity of radio communications processedVSAvoidrelevance of information
Core Design Contradiction:
Quantity of substanceVSLoss of information

Solution Approach 1:

The system extracts and displays only the subset of radio communications that are relevant to the specific aircraft by comparing speech recognition results against stored aircraft identifiers. This filters out communications directed to other aircraft, presenting only applicable information to the flight crew.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system applies different processing quality to different communications: full processing and display for relevant communications (those matching the aircraft's identifier), and filtering/out for irrelevant communications. This ensures high relevance quality for displayed information.

Inventive Principle:
Principle #3Local quality

2Ease of operation

If the system uses speech recognition to transcribe radio communications, then flight crew members benefit from transcription aids, but the transcriptions often include communications directed to other aircraft in the plurality

Engineering Contradiction:
Improvetranscription aid availabilityVSAvoidcommunication relevance
Core Design Contradiction:
Ease of operationVSLoss of information

Solution Approach 1:

The system introduces an intermediary filtering step between speech recognition transcription and display. This intermediary compares the transcribed communication against stored aircraft identifiers and only passes relevant communications to the display, eliminating irrelevant information while preserving the transcription aid benefit.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Quantity of substance

If the system displays all transcribed radio communications, then comprehensive information is available, but irrelevant communications clutter the display and reduce clarity for flight crews

Engineering Contradiction:
Improveinformation volumeVSAvoiddisplay clarity
Core Design Contradiction:
Quantity of substanceVSEase of operation

Solution Approach 1:

The system extracts only the relevant subset of transcribed communications by comparing against aircraft identifiers, removing irrelevant communications from the display. This maintains comprehensive information for relevant communications while eliminating clutter from unrelated communications.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP3336838B1Methods and apparatus for post-processing speech recognition results of received radio voice messages onboard an aircraft
Publication Date: 2024.01.31 HONEYWELL INTERNATIONAL INC
  • EP3336838B1 patent drawingFigure 1
  • EP3336838B1 patent drawingFigure 2
  • EP3336838B1 patent drawingFigure 3

AI summary

A method for displaying received radio voice messages onboard an aircraft is provided. The method post-processes, by at least one processor onboard the aircraft, a set of speech recognition (SR) hypothetical data to increase accuracy of an associated SR system, by: obtaining, by the at least one processor, secondary source data from a plurality of secondary sources; comparing, by the at least one processor, the set of SR hypothetical data to the secondary source data; and identifying, by the at least one processor, an aircraft tail number using the set of SR hypothetical data and the secondary source data; identifies, by the at least one processor, a subset of the received radio voice messages including the tail number; and presents, via a display device onboard the aircraft, the subset using distinguishing visual characteristics.