Skip to main content

Readback Error Detection by Automatic Speech Recognition and Understanding — Results of HAAWAII Project for Isavia’s Enroute Airspace

Paper ID

SIDs-2022-003

Conference

SESAR Innovation Days

Year

2022

Theme

Automatic Speech Recognition II

Project Name

SESAR 2020 ER4 project HAAWAII

Keywords:

air traffic control, Assistant Based Speech Recognition, Machine learning, Readback Error Detection, Speech Recognition, Speech Understanding

Authors

Hartmut Helmke, Karel Ondřej, Shruthi Shetty, Hörður Arilíusson, Teodor Simiganoschi, Matthias Kleinert, Oliver Ohneiser, Heiko Ehr, Juan Pablo Zuluaga and Pavel Smrz

DOI

Project Number

884287

Abstract

One of the crucial tasks of an air traffic controller (ATCo) is to evaluate pilot readbacks and to react in case of errors. Undetected readback errors, when not corrected by the ATCo, can have a dramatic impact on air traffic management (ATM) safety. Although they seldom occur, the benefits of even one prevented incident due to automatic readback error detection justify the efforts. The HAAWAII project uses automatic speech recognition and understanding (ASRU) to support the ATCo in this critical task. This paper presents for readback error detection approaches: a rule-based and a data-driven approach based on machine learning. The combination of both detects 81% of the readback error samples on real-life voice recordings from Isavia’s en-route airspace. Proof-of-concept trials with six ATCos from Isavia producing artificial, but challenging readback error samples resulted in a false alarm rate of 11% and a readback error detection rate of 80%. These results are based on Word Error Rates of 5% for ATCos and 10% for pilots, respectively.