How Does Pre-Trained Wav2Vec 2.0 Perform on Domain-Shifted ASR? an Extensive Benchmark on Air Traffic Control Communications

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F23%3APU149421" target="_blank" >RIV/00216305:26230/23:PU149421 - isvavai.cz</a>

Výsledek na webu

<a href="https://ieeexplore.ieee.org/document/10022724" target="_blank" >https://ieeexplore.ieee.org/document/10022724</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/SLT54892.2023.10022724" target="_blank" >10.1109/SLT54892.2023.10022724</a>

Jazyk výsledku

angličtina

Název v původním jazyce

How Does Pre-Trained Wav2Vec 2.0 Perform on Domain-Shifted ASR? an Extensive Benchmark on Air Traffic Control Communications

Popis výsledku v původním jazyce

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.

Název v anglickém jazyce

How Does Pre-Trained Wav2Vec 2.0 Perform on Domain-Shifted ASR? an Extensive Benchmark on Air Traffic Control Communications

Popis výsledku anglicky

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.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

R - Projekt Ramcoveho programu EK

Rok uplatnění

2023

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

IEEE Spoken Language Technology Workshop, SLT 2022 - Proceedings

ISBN

978-1-6654-7189-3

ISSN

—

e-ISSN

—

Počet stran výsledku

8

Strana od-do

205-212

Název nakladatele

IEEE Signal Processing Society

Místo vydání

Doha

Místo konání akce

Doha

Datum konání akce

9. 1. 2023

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000968851900028