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

Existing Self-Supervised Learning (SSL) models for speech typically processspeech signals at a fixed resolution of 20 milliseconds. This approachoverlooks the varying informational content present at different resolutions inspeech signals. In contrast, this paper aims to incorporate multi-resolutioninformation into speech self-supervised representation learning. We introduce aSSL model that leverages a hierarchical Transformer architecture, complementedby HuBERT-style masked prediction objectives, to process speech at multipleresolutions. Experimental results indicate that the proposed model not onlyachieves more efficient inference but also exhibits superior or comparableperformance to the original HuBERT model over various tasks. Specifically,significant performance improvements over the original HuBERT have beenobserved in fine-tuning experiments on the LibriSpeech speech recognitionbenchmark as well as in evaluations using the Speech Universal PERformanceBenchmark (SUPERB) and Multilingual SUPERB (ML-SUPERB).

Multi-resolution HuBERT: Multi-resolution Speech Self-Supervised Learning with Masked Unit Prediction