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

This paper addresses the observed performance gap between automatic speechrecognition (ASR) systems based on Long Short Term Memory (LSTM) neuralnetworks trained with the connectionist temporal classification (CTC) lossfunction and systems based on hybrid Deep Neural Networks (DNNs) trained withthe cross entropy (CE) loss function on domains with limited data. We stepthrough a number of experiments that show incremental improvements on abaseline EESEN toolkit based LSTM-CTC ASR system trained on the Librispeech100hr (train-clean-100) corpus. Our results show that with effectivecombination of data augmentation and regularization, a LSTM-CTC based systemcan exceed the performance of a strong Kaldi based baseline trained on the samedata.

Improving LSTM-CTC based ASR performance in domains with limited training data