Publication Details

Readback Error Detection by Automatic Speech Recognition and Understanding - Results of HAAWAII project for Isavia's Enroute Airspace

HELMKE Hartmut, ONDŘEJ Karel, SHETTY Shruthi, KLEINERT Matthias, OHNEISER Oliver, EHR Heiko, ZULUAGA-GOMEZ Juan and SMRŽ Pavel et al. Readback Error Detection by Automatic Speech Recognition and Understanding - Results of HAAWAII project for Isavia's Enroute Airspace. In: SESAR Innovation Days 2022. Budapest, 2022, pp. 1-9. Available from: https://www.sesarju.eu/sites/default/files/documents/sid/2022/paper_3.pdf
Czech title
Detekce chyb zpětného čtení pomocí automatickému rozpoznávání a porozumění řeči - Výsledky projektu HAAWAII pro vzdušný prostor řízený společností ISAVIA
Type
conference paper
Language
english
Authors
Helmke Hartmut (DLR)
Ondřej Karel, Ing. (DCGM FIT BUT)
Shetty Shruthi (DLR)
Kleinert Matthias (DLR)
Ohneiser Oliver (DLR)
Ehr Heiko (DLR)
Zuluaga-Gomez Juan (IDIAP)
and others
URL
Keywords

readback error detection, speech recognition, speech understanding, air traffic control, assistant based speech recognition, machine learning

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 Isavias 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. 

Published
2022
Pages
1-9
Proceedings
SESAR Innovation Days 2022
Conference
12th SESAR Innovation Days, Budapešť, HU
Place
Budapest, HU
BibTeX 
@INPROCEEDINGS{FITPUB12928,
   author = "Hartmut Helmke and Karel Ond\v{r}ej and Shruthi Shetty and Matthias Kleinert and Oliver Ohneiser and Heiko Ehr and Juan Zuluaga-Gomez and Pavel Smr\v{z} and et al.",
   title = "Readback Error Detection by Automatic Speech Recognition and Understanding - Results of HAAWAII project for Isavia's Enroute Airspace",
   pages = "1--9",
   booktitle = "SESAR Innovation Days 2022",
   year = 2022,
   location = "Budapest, HU",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/12928"
}
Back to top