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 Transformer based models, such as wav2vec 2.0 and HuBERT,have produced significant improvements over existing approaches to automaticspeech recognition (ASR). This is evident in the performance of the wav2vec 2.0based pretrained XLSR-53 model across many languages when fine-tuned withavailable labeled data. However, the performance from finetuning these modelscan be dependent on the amount of in-language or similar-to-in-language dataincluded in the pretraining dataset. In this paper we investigate continuedpretraining (CoPT) with unlabeled in-language audio data on the XLSR-53pretrained model in several low-resource languages. CoPT is morecomputationally efficient than semi-supervised training (SST), the standardapproach of utilizing unlabeled data in ASR, since it omits the need forpseudo-labeling of the unlabeled data. We show CoPT results in word error rates(WERs), equal to or slightly better than using SST. In addition, we show thatusing the CoPT model for pseudo-labeling, and using these labels in SST,results in further improvements in WER.

Improving Low-Resource Speech Recognition with Pretrained Speech Models: Continued Pretraining vs. Semi-Supervised Training