On-Board Speech Recognition for Aviation Communication

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current air traffic control speech recognition systems face challenges in achieving high accuracy due to noisy environments, professional terminology, regional differences, and accent variations, leading to potential miscommunication and safety risks during aircraft navigation.

Innovation Solution

An on-board information assistance system that captures real-time air-ground communication speech, segments speech data based on information entropy, and uses heterogeneous speech processing modules to convert speech into text, displaying both speech and text simultaneously to improve understanding and accuracy, with the text data being used for flight guidance instructions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If ordinary speech recognition products are used for air traffic control dialogue recognition, then device complexity is reduced, but recognition accuracy deteriorates due to noisy environments, professional terminology, regional differences, and accent variations

Engineering Contradiction:
Improvespeech recognition accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments speech data based on information entropy to divide continuous speech into discrete segments for processing. This segmentation approach allows the system to handle complex speech recognition tasks by breaking them into manageable units, improving accuracy without proportionally increasing overall system complexity

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary processing layer that converts speech to text and then processes the text information separately. This intermediary text representation acts as a bridge between the noisy speech input and the navigation system, filtering out noise and improving recognition accuracy while maintaining manageable system complexity

Inventive Principle:
Principle #24Intermediary (Mediator)

2Loss of information

If speech recognition is performed on real-time air-ground communication, then information availability is improved, but reliability deteriorates due to potential mishearing and omission of information

Engineering Contradiction:
Improveinformation completenessVSAvoidcommunication reliability
Core Design Contradiction:
Loss of informationVSReliability

Solution Approach 1:

The patent implements a feedback mechanism where the system processes speech information through multiple stages including speech-to-text conversion, text processing, and validation. This multi-stage feedback process allows the system to detect and correct potential errors in speech recognition, improving reliability while maintaining information completeness

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent performs preliminary processing of speech data by converting it to text format before final interpretation and action. This preliminary text conversion allows for error detection and correction before the information is used for navigation decisions, enhancing reliability while preserving information completeness

Inventive Principle:
Principle #10Preliminary action

3Speed

If speech data is processed in real-time during flight operations, then response speed is improved, but measurement precision deteriorates due to noisy environments and communication link interference

Engineering Contradiction:
Improveresponse speedVSAvoidspeech recognition precision
Core Design Contradiction:
SpeedVSMeasurement precision

Solution Approach 1:

The patent segments speech data based on information entropy, allowing real-time processing of divided speech units rather than waiting for complete sentences. This segmentation enables faster response while each segment is processed with sufficient precision to maintain overall recognition accuracy

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent creates text copies of speech data that can be processed independently from the original audio signal. This copying approach allows real-time processing of the text representation while the original speech can be reprocessed if needed, improving response speed without sacrificing precision

Inventive Principle:
Principle #26Copying

Data Source

PatentEP4044179B1On-board information assisting system and method
Publication Date: 2024.11.13 BEIJING AERONAUTIC SCI & TECH RES INST OF COMAC
  • EP4044179B1 patent drawingFigure 1
  • EP4044179B1 patent drawingFigure 2
  • EP4044179B1 patent drawingFigure 3

AI summary

An on-board information assistance system and method include: obtaining, by an obtaining module (10), speech information, converting, by speech processing modules (21) and (22), the speech information into text information, and displaying the speech information and the text information on a man-machine interactive interface (40) by means of an electronic flight bag (31) and an instruction recording system (32); or, receiving, by a navigation system (30), a navigation instruction in the text information for flight guidance. The on-board information assistance system and method relates to the field of aviation and is used to assist pilots in listening, replaying, and querying air traffic control information during flight missions, improve the convenience and accuracy of obtaining control instructions, reduce communication costs and provide safety guarantees for aircraft navigation.