English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Self-Supervised Learning (SSL) has allowed leveraging large amounts ofunlabeled speech data to improve the performance of speech recognition modelseven with small annotated datasets. Despite this, speech SSL representationsmay fail while facing an acoustic mismatch between the pretraining and targetdatasets. To address this issue, we propose a novel supervised domainadaptation method, designed for cases exhibiting such a mismatch in acousticdomains. It consists in applying properly calibrated data augmentations on alarge clean dataset, bringing it closer to the target domain, and using it aspart of an initial fine-tuning stage. Augmentations are automatically selectedthrough the minimization of a conditional-dependence estimator, based on thetarget dataset. The approach is validated during an oracle experiment withcontrolled distortions and on two amateur-collected low-resource domains,reaching better performances compared to the baselines in both cases.

Automatic Data Augmentation for Domain Adapted Fine-Tuning of Self-Supervised Speech Representations