Robust Wav2vec 2.0: Analyzing Domain Shift In Self-supervised Pre-training

Abstract

Self-supervised learning of speech representations has been a very active research area but most work is focused on a single domain such as read audio books for which there exist large quantities of labeled and unlabeled data. In this paper, we explore more general setups where the domain of the unlabeled data for pre-training data differs from the domain of the labeled data for fine-tuning, which in turn may differ from the test data domain. Our experiments show that using target domain data during pre-training leads to large performance improvements across a variety of setups. On a large-scale competitive setup, we show that pre-training on unlabeled in-domain data reduces the gap between models trained on in-domain and out-of-domain labeled data by 66%-73%. This has obvious practical implications since it is much easier to obtain unlabeled target domain data than labeled data. Moreover, we find that pre-training on multiple domains improves generalization performance on domains not

Stats

• citations137
• S2 citations—
• github stars32213
• HF likes0
• heat score25.07
• arxiv keyhsu2021robust

Related papers

• A Noise-robust Self-supervised Pre-training Model Based Speech Representation Learning For Automatic Speech Recognition (2022)11.19
• Boosting Cross-domain Speech Recognition With Self-supervision (2022)0.00
• Measuring The Impact Of Individual Domain Factors In Self-supervised Pre-training (2022)5.24
• Analyzing The Robustness Of Unsupervised Speech Recognition (2021)7.81
• Ccc-wav2vec 2.0: Clustering Aided Cross Contrastive Self-supervised Learning Of Speech Representations (2022)7.81
• Layer-wise Analysis Of A Self-supervised Speech Representation Model (2021)17.07
• Wav2vec-s: Semi-supervised Pre-training For Low-resource ASR (2021)7.50
• Automatic Data Augmentation For Domain Adapted Fine-tuning Of Self-supervised Speech Representations (2023)0.00