The development, evaluation and application of an aviation radiotelephony specialised technical vocabulary list

Drayton, Jenny; Coxhead, Averil

doi:10.1016/j.esp.2022.10.001

Cited by 9 publications

(3 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Pilot repetition instructions require responding to the control instructions based on ATC rules without losing any crucial information. According to ATC rules (Drayton and Coxhead, 2023 ), ATCO instructions must start with the aircraft identification (ACID) to specify the communicating aircraft, while pilot repetitions should end with their ACID to differentiate them from ATCO instructions. Based on the characteristics of the generated repetitions mentioned above, using only the ROUGE evaluation metric cannot comprehensively assess the model's performance.…”

Section: Methodsmentioning

confidence: 99%

Research on automatic pilot repetition generation method based on deep reinforcement learning

Pan,

Jiang,

et al. 2023

Front. Neurorobot.

View full text Add to dashboard Cite

Using computers to replace pilot seats in air traffic control (ATC) simulators is an effective way to improve controller training efficiency and reduce training costs. To achieve this, we propose a deep reinforcement learning model, RoBERTa-RL (RoBERTa with Reinforcement Learning), for generating pilot repetitions. RoBERTa-RL is based on the pre-trained language model RoBERTa and is optimized through transfer learning and reinforcement learning. Transfer learning is used to address the issue of scarce data in the ATC domain, while reinforcement learning algorithms are employed to optimize the RoBERTa model and overcome the limitations in model generalization caused by transfer learning. We selected a real-world area control dataset as the target task training and testing dataset, and a tower control dataset generated based on civil aviation radio land-air communication rules as the test dataset for evaluating model generalization. In terms of the ROUGE evaluation metrics, RoBERTa-RL achieved significant results on the area control dataset with ROUGE-1, ROUGE-2, and ROUGE-L scores of 0.9962, 0.992, and 0.996, respectively. On the tower control dataset, the scores were 0.982, 0.954, and 0.982, respectively. To overcome the limitations of ROUGE in this field, we conducted a detailed evaluation of the proposed model architecture using keyword-based evaluation criteria for the generated repetition instructions. This evaluation criterion calculates various keyword-based metrics based on the segmented results of the repetition instruction text. In the keyword-based evaluation criteria, the constructed model achieved an overall accuracy of 98.8% on the area control dataset and 81.8% on the tower control dataset. In terms of generalization, RoBERTa-RL improved accuracy by 56% compared to the model before improvement and achieved a 47.5% improvement compared to various comparative models. These results indicate that employing reinforcement learning strategies to enhance deep learning algorithms can effectively mitigate the issue of poor generalization in text generation tasks, and this approach holds promise for future application in other related domains.

show abstract

Section: Methodsmentioning

confidence: 99%

Research on automatic pilot repetition generation method based on deep reinforcement learning

Pan,

Jiang,

et al. 2023

Front. Neurorobot.

View full text Add to dashboard Cite

show abstract

“…Therefore, a Semantic WER (SWER) is proposed to address the limitations of WER in terms of usability. The ICAO standard mentioned in [45] states that ATCO instructions must start with aircraft identification (ACID) to identify the communicating aircraft, and pilot instructions must end with their ACID to distinguish them from ATCO instructions, indicating that control instructions have significant structural features. A correct control instruction contains several essential elements, such as the call sign, action instruction, and action parameters.…”

Section: Evaluation Metrics and Methodsmentioning

confidence: 99%

A Robust Conformer-Based Speech Recognition Model for Mandarin Air Traffic Control

Jiang,

Pan,

Zhang

et al. 2023

CMC

View full text Add to dashboard Cite

This study aims to address the deviation in downstream tasks caused by inaccurate recognition results when applying Automatic Speech Recognition (ASR) technology in the Air Traffic Control (ATC) field. This paper presents a novel cascaded model architecture, namely Conformer-CTC/Attention-T5 (CCAT), to build a highly accurate and robust ATC speech recognition model. To tackle the challenges posed by noise and fast speech rate in ATC, the Conformer model is employed to extract robust and discriminative speech representations from raw waveforms. On the decoding side, the Attention mechanism is integrated to facilitate precise alignment between input features and output characters. The Text-To-Text Transfer Transformer (T5) language model is also introduced to handle particular pronunciations and code-mixing issues, providing more accurate and concise textual output for downstream tasks. To enhance the model's robustness, transfer learning and data augmentation techniques are utilized in the training strategy. The model's performance is optimized by performing hyperparameter tunings, such as adjusting the number of attention heads, encoder layers, and the weights of the loss function. The experimental results demonstrate the significant contributions of data augmentation, hyperparameter tuning, and error correction models to the overall model performance. On the Our ATC Corpus dataset, the proposed model achieves a Character Error Rate (CER) of 3.44%, representing a 3.64% improvement compared to the baseline model. Moreover, the effectiveness of the proposed model is validated on two publicly available datasets. On the AISHELL-1 dataset, the CCAT model achieves a CER of 3.42%, showcasing a 1.23% improvement over the baseline model. Similarly, on the LibriSpeech dataset, the CCAT model achieves a Word Error Rate (WER) of 5.27%, demonstrating a performance improvement of 7.67% compared to the baseline model. Additionally, this paper proposes an evaluation criterion for assessing the robustness of ATC speech recognition systems. In robustness evaluation experiments based on this criterion, the proposed model demonstrates a performance improvement of 22% compared to the baseline model.

show abstract

“…Understanding and using this vocabulary shows that these learners belong to a particular community (Coxhead, 2012). The main research focus in this area includes identifying specialized vocabulary (e.g., Chung & Nation, 2003;Coxhead, 2018), creating word lists (e.g., Coxhead et al, 2019;Fraser, 2007;Valipouri & Nassaji, 2013;Ward, 2009), evaluating word lists (e.g., Drayton & Coxhead, 2023;Nguyen & Coxhead, 2022;Nation, 2016), and multi-word units (e.g., Benson & Coxhead, 2022;Wood & Appel, 2014). Medicine is well known for its large amount of technical vocabulary, mostly borrowed from Greek and Latin and some with little morphological adaptation (e.g., diabetes, embolus) (Ferguson, 2012).…”

Section: Linguistic Inquiries Into Esp/emp and Their Connection To Te...mentioning

confidence: 99%

Investigating English for Medical Purposes (EMP) Coursebooks in China: Insights from Learner and Teacher Questionnaires and Coursebook Analysis

View full text Add to dashboard Cite

English for Medical Purposes (EMP) education is growing rapidly in China, with over 100 medical universities or faculties offering EMP courses. The overarching aim of this research is to investigate how EMP coursebooks can effectively support Chinese tertiary EMP learners and teachers through a mixed method approach. Study 1 involved developing, validating, and delivering a learner questionnaire and a teacher questionnaire to explore the use, evaluation, and needs regarding EMP coursebooks. Study 2 focused on one commonly used EMP coursebook identified from Study 1 by applying and adapting Littlejohn’s (2011) framework to analyze its content. The integration of the two studies allowed for an in-depth investigation into the effectiveness of EMP coursebooks in supporting Chinese tertiary EMP learners and teachers.Study 1 began with determining content areas and drafting the questionnaires in English, followed by examining the representativeness and relevance of the content areas and questionnaire items. The readability of both the English questionnaires and their Chinese translations was assessed before proceeding to the piloting stage. The learner questionnaire underwent two rounds of Exploratory Factor Analysis (EFA) and one Confirmatory Factor Analysis (CFA). The EFAs revealed that the underlying structure of the learner responses did not match the theoretical structure of the questionnaire well. The structure identified from the second EFA was modified to better fit the data in the CFA. The modified structure of the learner questionnaire consisted of five variables. Two variables were related to coursebook use: (1) teacher-supervised coursebook preview and review and (2) learners’ self-directed coursebook use. One variable concerned coursebook evaluation: reflection of how EMP is actually used in the field of medicine by providing texts and topics medical students may encounter in studies and future work. The remaining two variables involved learners’ needs from the coursebooks: (1) scaffolding, such as providing answer keys and tips for completing activities, and (2) interactive activities like pair or group work.Several key findings emerged from the descriptive, inferential, and qualitative content analyses of the learner questionnaire data (n = 522). First, learners often used their EMP coursebooks in a self-directed way. Second, they agreed that their coursebooks effectively reflected how EMP is actually used in the field of medicine. Third, they needed scaffolding and interactive activities in the coursebooks. Fourth, they needed great support in learning medical vocabulary and paid extra attention to specialization of medical content, meta-language, organization of content, and use of images in the coursebooks. The statistical and qualitative content analyses of the teacher responses (n = 51) revealed that the teachers rarely created their own materials but frequently adapted their EMP coursebooks. They agreed on the effectiveness of their coursebooks in reflecting how EMP is actually used in the field of medicine. Their strongest need regarding EMP coursebooks was guidance on vocabulary use, followed by supporting resources on medical content. The teacher questionnaire did not undergo EFA and CFA.In Study 2, one EMP coursebook was analyzed using an adapted version of Littlejohn’s (2011) three-level framework: (1) objective description; (2) task analysis in terms of process, mode of participation, and content; and (3) inference about underlying principles. The analysis revealed that the coursebook provided sufficient supporting resources to scaffold EMP learning and teaching, including video scripts, Chinese translations of the texts, and answer keys. Most coursebook images supported learners’ understanding of texts. The limitations included (1) a lack of guidance on medical vocabulary frequency and register, (2) threats to the authenticity of original texts through adaptations, (3) absence of some common medical text types, such as case reports and medical records, and (4) little attention to discoursal features, such as grammatical features and genre conventions.This research provides guidance for coursebook designers in designing EMP coursebooks that can effectively support EMP education in China. Findings can raise EMP teachers’ awareness of addressing deficiencies in their coursebooks by adapting the materials effectively. Moreover, this research highlights the importance of using rigorous approaches to develop questionnaires with sufficient reliability and validity.

show abstract

The development, evaluation and application of an aviation radiotelephony specialised technical vocabulary list

Cited by 9 publications

References 18 publications

Research on automatic pilot repetition generation method based on deep reinforcement learning

Research on automatic pilot repetition generation method based on deep reinforcement learning

A Robust Conformer-Based Speech Recognition Model for Mandarin Air Traffic Control

Investigating English for Medical Purposes (EMP) Coursebooks in China: Insights from Learner and Teacher Questionnaires and Coursebook Analysis

Contact Info

Product

Resources

About