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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Advancements in monaural speech enhancement (SE) techniques have greatlyimproved the perceptual quality of speech. However, integrating thesetechniques into automatic speech recognition (ASR) systems has not yielded theexpected performance gains, primarily due to the introduction of distortionsduring the SE process. In this paper, we propose a novel approach calledFAT-HuBERT, which leverages distortion-invariant self-supervised learning (SSL)to enhance the robustness of ASR. To address the distortions introduced by theSE frontends, we introduce layer-wise fusion modules that incorporate featuresextracted from both observed noisy signals and enhanced signals. Duringtraining, the SE frontend is randomly selected from a pool of models. Weevaluate the performance of FAT-HuBERT on simulated noisy speech generated fromLibriSpeech as well as real-world noisy speech from the CHiME-4 1-channeldataset. The experimental results demonstrate a significant relative reductionin word error rate (WER).

FAT-HuBERT: Front-end Adaptive Training of Hidden-unit BERT for Distortion-Invariant Robust Speech Recognition