How Does Pre-trained Wav2Vec 2.0 Perform on Domain Shifted ASR? An Extensive Benchmark on Air Traffic Control Communications
This work addresses domain shift in ASR for air traffic control, an incremental study focusing on a specific domain.
The paper tackled the problem of domain shift in automatic speech recognition by evaluating pre-trained Wav2Vec 2.0 and XLS-R models on air traffic control communications, achieving relative word error rate reductions of 20% to 40% compared to hybrid baselines with less labeled data.
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 and 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.