A Context-Aware Language Model to Improve the Speech Recognition in Air Traffic Control
Recognizing isolated digits of the flight callsign is an important and challenging task for automatic speech recognition (ASR) in air traffic control (ATC). Fortunately, the flight callsign is a kind of prior ATC knowledge and is available from dynamic contextual information. In this work, we attemp...
Guardado en:
Autores principales: | , , , , |
---|---|
Formato: | article |
Lenguaje: | EN |
Publicado: |
MDPI AG
2021
|
Materias: | |
Acceso en línea: | https://doaj.org/article/d5c3f6acd8874e9bb51a0fe02f3ccc33 |
Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
id |
oai:doaj.org-article:d5c3f6acd8874e9bb51a0fe02f3ccc33 |
---|---|
record_format |
dspace |
spelling |
oai:doaj.org-article:d5c3f6acd8874e9bb51a0fe02f3ccc332021-11-25T15:57:47ZA Context-Aware Language Model to Improve the Speech Recognition in Air Traffic Control10.3390/aerospace81103482226-4310https://doaj.org/article/d5c3f6acd8874e9bb51a0fe02f3ccc332021-11-01T00:00:00Zhttps://www.mdpi.com/2226-4310/8/11/348https://doaj.org/toc/2226-4310Recognizing isolated digits of the flight callsign is an important and challenging task for automatic speech recognition (ASR) in air traffic control (ATC). Fortunately, the flight callsign is a kind of prior ATC knowledge and is available from dynamic contextual information. In this work, we attempt to utilize this prior knowledge to improve the performance of the callsign identification by integrating it into the language model (LM). The proposed approach is named context-aware language model (CALM), which can be applied for both the ASR decoding and rescoring phase. The proposed model is implemented with an encoder–decoder architecture, in which an extra context encoder is proposed to consider the contextual information. A shared embedding layer is designed to capture the correlations between the ASR text and contextual information. The context attention is introduced to learn discriminative representations to support the decoder module. Finally, the proposed approach is validated with an end-to-end ASR model on a multilingual real-world corpus (ATCSpeech). Experimental results demonstrate that the proposed CALM outperforms other baselines for both the ASR and callsign identification task, and can be practically migrated to a real-time environment.Dongyue GuoZichen ZhangPeng FanJianwei ZhangBo YangMDPI AGarticlelanguage modelautomatic speech recognitionair traffic controldynamic contextual informationMotor vehicles. Aeronautics. AstronauticsTL1-4050ENAerospace, Vol 8, Iss 348, p 348 (2021) |
institution |
DOAJ |
collection |
DOAJ |
language |
EN |
topic |
language model automatic speech recognition air traffic control dynamic contextual information Motor vehicles. Aeronautics. Astronautics TL1-4050 |
spellingShingle |
language model automatic speech recognition air traffic control dynamic contextual information Motor vehicles. Aeronautics. Astronautics TL1-4050 Dongyue Guo Zichen Zhang Peng Fan Jianwei Zhang Bo Yang A Context-Aware Language Model to Improve the Speech Recognition in Air Traffic Control |
description |
Recognizing isolated digits of the flight callsign is an important and challenging task for automatic speech recognition (ASR) in air traffic control (ATC). Fortunately, the flight callsign is a kind of prior ATC knowledge and is available from dynamic contextual information. In this work, we attempt to utilize this prior knowledge to improve the performance of the callsign identification by integrating it into the language model (LM). The proposed approach is named context-aware language model (CALM), which can be applied for both the ASR decoding and rescoring phase. The proposed model is implemented with an encoder–decoder architecture, in which an extra context encoder is proposed to consider the contextual information. A shared embedding layer is designed to capture the correlations between the ASR text and contextual information. The context attention is introduced to learn discriminative representations to support the decoder module. Finally, the proposed approach is validated with an end-to-end ASR model on a multilingual real-world corpus (ATCSpeech). Experimental results demonstrate that the proposed CALM outperforms other baselines for both the ASR and callsign identification task, and can be practically migrated to a real-time environment. |
format |
article |
author |
Dongyue Guo Zichen Zhang Peng Fan Jianwei Zhang Bo Yang |
author_facet |
Dongyue Guo Zichen Zhang Peng Fan Jianwei Zhang Bo Yang |
author_sort |
Dongyue Guo |
title |
A Context-Aware Language Model to Improve the Speech Recognition in Air Traffic Control |
title_short |
A Context-Aware Language Model to Improve the Speech Recognition in Air Traffic Control |
title_full |
A Context-Aware Language Model to Improve the Speech Recognition in Air Traffic Control |
title_fullStr |
A Context-Aware Language Model to Improve the Speech Recognition in Air Traffic Control |
title_full_unstemmed |
A Context-Aware Language Model to Improve the Speech Recognition in Air Traffic Control |
title_sort |
context-aware language model to improve the speech recognition in air traffic control |
publisher |
MDPI AG |
publishDate |
2021 |
url |
https://doaj.org/article/d5c3f6acd8874e9bb51a0fe02f3ccc33 |
work_keys_str_mv |
AT dongyueguo acontextawarelanguagemodeltoimprovethespeechrecognitioninairtrafficcontrol AT zichenzhang acontextawarelanguagemodeltoimprovethespeechrecognitioninairtrafficcontrol AT pengfan acontextawarelanguagemodeltoimprovethespeechrecognitioninairtrafficcontrol AT jianweizhang acontextawarelanguagemodeltoimprovethespeechrecognitioninairtrafficcontrol AT boyang acontextawarelanguagemodeltoimprovethespeechrecognitioninairtrafficcontrol AT dongyueguo contextawarelanguagemodeltoimprovethespeechrecognitioninairtrafficcontrol AT zichenzhang contextawarelanguagemodeltoimprovethespeechrecognitioninairtrafficcontrol AT pengfan contextawarelanguagemodeltoimprovethespeechrecognitioninairtrafficcontrol AT jianweizhang contextawarelanguagemodeltoimprovethespeechrecognitioninairtrafficcontrol AT boyang contextawarelanguagemodeltoimprovethespeechrecognitioninairtrafficcontrol |
_version_ |
1718413408803487744 |