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

Recognizing code-switched speech is challenging for Automatic SpeechRecognition (ASR) for a variety of reasons, including the lack of code-switchedtraining data. Recently, we showed that monolingual ASR systems fine-tuned oncode-switched data deteriorate in performance on monolingual speechrecognition, which is not desirable as ASR systems deployed in multilingualscenarios should recognize both monolingual and code-switched speech with highaccuracy. Our experiments indicated that this loss in performance could bemitigated by using certain strategies for fine-tuning and regularization,leading to improvements in both monolingual and code-switched ASR. In thiswork, we present further improvements over our previous work by using domainadversarial learning to train task agnostic models. We evaluate theclassification accuracy of an adversarial discriminator and show that it canlearn shared layer parameters that are task agnostic. We train end-to-end ASRsystems starting with a pooled model that uses monolingual and code-switcheddata along with the adversarial discriminator. Our proposed technique leads toreductions in Word Error Rates (WER) in monolingual and code-switched test setsacross three language pairs.

Learning not to Discriminate: Task Agnostic Learning for Improving Monolingual and Code-switched Speech Recognition