Publication Details
How Does Pre-Trained Wav2Vec 2.0 Perform on Domain-Shifted ASR? an Extensive Benchmark on Air Traffic Control Communications
Prasad Amrutha (DCGM)
NIGMATULINA, I.
Sarfjoo Seyyed Saeed
Motlíček Petr, doc. Ing., Ph.D. (DCGM)
KLEINERT, M.
HELMKE, H.
OHNEISER, O.
ZHAN, Q.
Automatic speech recognition, Wav2Vec 2.0, self-supervised pre-training, air traffic control communications.
Recent work on self-supervised pre-training focus on leveraging
large-scale unlabeled speech data to build robust end-to-end (E2E)
acoustic models (AM) that can be later fine-tuned on downstream
tasks e.g., automatic speech recognition (ASR). Yet, few works
investigated the impact on performance when the data properties
substantially differ between the pre-training and fine-tuning phases,
termed domain shift. We target this scenario by analyzing the robustness
of Wav2Vec 2.0 and XLS-R models on downstream ASR
for a completely unseen domain, air traffic control (ATC) communications.
We benchmark these two models on several open-source
and challenging ATC databases with signal-to-noise ratio between 5
to 20 dB. Relative word error rate (WER) reductions between 20%
to 40% are obtained in comparison to hybrid-based ASR baselines
by only fine-tuning E2E acoustic models with a smaller fraction of
labeled data. We analyze WERs on the low-resource scenario and
gender bias carried by one ATC dataset.
@inproceedings{BUT185194,
author="ZULUAGA-GOMEZ, J. and PRASAD, A. and NIGMATULINA, I. and SARFJOO, S. and MOTLÍČEK, P. and KLEINERT, M. and HELMKE, H. and OHNEISER, O. and ZHAN, Q.",
title="How Does Pre-Trained Wav2Vec 2.0 Perform on Domain-Shifted ASR? an Extensive Benchmark on Air Traffic Control Communications",
booktitle="IEEE Spoken Language Technology Workshop, SLT 2022 - Proceedings",
year="2023",
pages="205--212",
publisher="IEEE Signal Processing Society",
address="Doha",
doi="10.1109/SLT54892.2023.10022724",
isbn="978-1-6654-7189-3",
url="https://ieeexplore.ieee.org/document/10022724"
}