Towards an Accent-Robust Approach for ATC Communications Transcription

Jahchan, Nataly; Barbier, Florentin; Gita, Ariyanidevi Dharma; Khelif, Khaled; Delpech, Estelle

doi:10.21437/interspeech.2021-333

Cited by 2 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This experiment focuses on evaluating the degree of accent in ATC speech, and observing and measuring the effect of ATC speech with different degrees of accent on the recognition accuracy performance of speech recognition models. According to a research study (Jahchan et al, 2021 ), multi-language inter-switching tends to produce more pronounced accents than geographic switching, and individual languages have their own pronunciation habits. For example, civil aviation pilots in various countries are generally familiar with the English pronunciation of the destination country in advance before executing international flights, so that they can quickly make corresponding feedback in the first time after receiving the speech control instructions to ensure the absolute safety of aviation operations (Romero-Rivas et al, 2015 ).…”

Section: Experimental Results and Validationmentioning

confidence: 99%

Assessment and analysis of accents in air traffic control speech: a fusion of deep learning and information theory

Pan,

Zhang,

Zhang

et al. 2024

Front. Neurorobot.

View full text Add to dashboard Cite

IntroductionEnhancing the generalization and reliability of speech recognition models in the field of air traffic control (ATC) is a challenging task. This is due to the limited storage, difficulty in acquisition, and high labeling costs of ATC speech data, which may result in data sample bias and class imbalance, leading to uncertainty and inaccuracy in speech recognition results. This study investigates a method for assessing the quality of ATC speech based on accents. Different combinations of data quality categories are selected according to the requirements of different model application scenarios to address the aforementioned issues effectively.MethodsThe impact of accents on the performance of speech recognition models is analyzed, and a fusion feature phoneme recognition model based on prior text information is constructed to identify phonemes of speech uttered by speakers. This model includes an audio encoding module, a prior text encoding module, a feature fusion module, and fully connected layers. The model takes speech and its corresponding prior text as input and outputs a predicted phoneme sequence of the speech. The model recognizes accented speech as phonemes that do not match the transcribed phoneme sequence of the actual speech text and quantitatively evaluates the accents in ATC communication by calculating the differences between the recognized phoneme sequence and the transcribed phoneme sequence of the actual speech text. Additionally, different levels of accents are input into different types of speech recognition models to analyze and compare the recognition accuracy of the models.ResultExperimental results show that, under the same experimental conditions, the highest impact of different levels of accents on speech recognition accuracy in ATC communication is 26.37%.DiscussionThis further demonstrates that accents affect the accuracy of speech recognition models in ATC communication and can be considered as one of the metrics for evaluating the quality of ATC speech.

show abstract

Section: Experimental Results and Validationmentioning

confidence: 99%

Assessment and analysis of accents in air traffic control speech: a fusion of deep learning and information theory

Pan,

Zhang,

Zhang

et al. 2024

Front. Neurorobot.

View full text Add to dashboard Cite

show abstract

“…Although the hybrid approach combines different algorithms and techniques, challenges in ASR systems still exist. To address issues such as poor signal quality from VHF communication, ambiguity in commands and instruction values, or the use of nonstandard phraseology and mispronunciation in different accents (native and non-native speakers) [11], a new approach based on utilizing contextual information is introduced to improve the performance and accuracy of ASR in ATC as a post-processing approach based on syntactic, semantic and pragmatic analysis.…”

Section: Asr In Atcmentioning

confidence: 99%

Toward Effective Aircraft Call Sign Detection Using Fuzzy String-Matching between ASR and ADS-B Data

Kasttet,

Lyhyaoui,

Zbakh

et al. 2023

Aerospace

View full text Add to dashboard Cite

Recently, artificial intelligence and data science have witnessed dramatic progress and rapid growth, especially Automatic Speech Recognition (ASR) technology based on Hidden Markov Models (HMMs) and Deep Neural Networks (DNNs). Consequently, new end-to-end Recurrent Neural Network (RNN) toolkits were developed with higher speed and accuracy that can often achieve a Word Error Rate (WER) below 10%. These toolkits can nowadays be deployed, for instance, within aircraft cockpits and Air Traffic Control (ATC) systems in order to identify aircraft and display recognized voice messages related to flight data, especially for airports not equipped with radar. Hence, the performance of air traffic controllers and pilots can ultimately be improved by reducing workload and stress and enforcing safety standards. Our experiment conducted at Tangier’s International Airport ATC aimed to build an ASR model that is able to recognize aircraft call signs in a fast and accurate way. The acoustic and linguistic models were trained on the Ibn Battouta Speech Corpus (IBSC), resulting in an unprecedented speech dataset with approved transcription that includes real weather aerodrome observation data and flight information with a call sign captured by an ADS-B receiver. All of these data were synchronized with voice recordings in a structured format. We calculated the WER to evaluate the model’s accuracy and compared different methods of dataset training for model building and adaptation. Despite the high interference in the VHF radio communication channel and fast-speaking conditions that increased the WER level to 20%, our standalone and low-cost ASR system with a trained RNN model, supported by the Deep Speech toolkit, was able to achieve call sign detection rate scores up to 96% in air traffic controller messages and 90% in pilot messages while displaying related flight information from ADS-B data using the Fuzzy string-matching algorithm.

show abstract

Towards an Accent-Robust Approach for ATC Communications Transcription

Cited by 2 publications

References 0 publications

Assessment and analysis of accents in air traffic control speech: a fusion of deep learning and information theory

Assessment and analysis of accents in air traffic control speech: a fusion of deep learning and information theory

Toward Effective Aircraft Call Sign Detection Using Fuzzy String-Matching between ASR and ADS-B Data

Contact Info

Product

Resources

About