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

The recently proposed conformer architecture has been successfully used forend-to-end automatic speech recognition (ASR) architectures achievingstate-of-the-art performance on different datasets. To our best knowledge, theimpact of using conformer acoustic model for hybrid ASR is not investigated. Inthis paper, we present and evaluate a competitive conformer-based hybrid modeltraining recipe. We study different training aspects and methods to improveword-error-rate as well as to increase training speed. We apply timedownsampling methods for efficient training and use transposed convolutions toupsample the output sequence again. We conduct experiments on Switchboard 300hdataset and our conformer-based hybrid model achieves competitive resultscompared to other architectures. It generalizes very well on Hub5'01 test setand outperforms the BLSTM-based hybrid model significantly.

Conformer-based Hybrid ASR System for Switchboard Dataset