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

End-to-end (E2E) spoken language understanding (SLU) systems can infer thesemantics of a spoken utterance directly from an audio signal. However,training an E2E system remains a challenge, largely due to the scarcity ofpaired audio-semantics data. In this paper, we treat an E2E system as amulti-modal model, with audio and text functioning as its two modalities, anduse a cross-modal latent space (CMLS) architecture, where a shared latent spaceis learned between the `acoustic' and `text' embeddings. We propose usingdifferent multi-modal losses to explicitly guide the acoustic embeddings to becloser to the text embeddings, obtained from a semantically powerfulpre-trained BERT model. We train the CMLS model on two publicly available E2Edatasets, across different cross-modal losses and show that our proposedtriplet loss function achieves the best performance. It achieves a relativeimprovement of 1.4% and 4% respectively over an E2E model without a cross-modalspace and a relative improvement of 0.7% and 1% over a previously publishedCMLS model using $L_2$ loss. The gains are higher for a smaller, morecomplicated E2E dataset, demonstrating the efficacy of using an efficientcross-modal loss function, especially when there is limited E2E training dataavailable.

Tie Your Embeddings Down: Cross-Modal Latent Spaces for End-to-end Spoken Language Understanding