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

In this work, we investigate if the wav2vec 2.0 self-supervised pretraininghelps mitigate the overfitting issues with connectionist temporalclassification (CTC) training to reduce its performance gap with flat-startlattice-free MMI (E2E-LFMMI) for automatic speech recognition with limitedtraining data. Towards that objective, we use the pretrained wav2vec 2.0 BASEmodel and fine-tune it on three different datasets including out-of-domain(Switchboard) and cross-lingual (Babel) scenarios. Our results show that forsupervised adaptation of the wav2vec 2.0 model, both E2E-LFMMI and CTC achievesimilar results; significantly outperforming the baselines trained only withsupervised data. Fine-tuning the wav2vec 2.0 model with E2E-LFMMI and CTC weobtain the following relative WER improvements over the supervised baselinetrained with E2E-LFMMI. We get relative improvements of 40% and 44% on theclean-set and 64% and 58% on the test set of Librispeech (100h) respectively.On Switchboard (300h) we obtain relative improvements of 33% and 35%respectively. Finally, for Babel languages, we obtain relative improvements of26% and 23% on Swahili (38h) and 18% and 17% on Tagalog (84h) respectively.

Comparing CTC and LFMMI for out-of-domain adaptation of wav2vec 2.0 acoustic model