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

Recent years have witnessed a boom in self-supervised learning (SSL) invarious areas including speech processing. Speech based SSL models presentpromising performance in a range of speech related tasks. However, the trainingof SSL models is computationally expensive and a common practice is tofine-tune a released SSL model on the specific task. It is essential to useconsistent front-end input during pre-training and fine-tuning. Thisconsistency may introduce potential issues when the optimal front-end is notthe same as that used in pre-training. In this paper, we propose a simple buteffective front-end adapter to address this front-end discrepancy. Byminimizing the distance between the outputs of different front-ends, thefilterbank feature (Fbank) can be compatible with SSL models which arepre-trained with waveform. The experiment results demonstrate the effectivenessof our proposed front-end adapter on several popular SSL models for the speechrecognition task.

Front-End Adapter: Adapting Front-End Input of Speech based Self-Supervised Learning for Speech Recognition