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 Spoken language understanding (SLU), a natural solution is concatenatingpre-trained speech models (e.g. HuBERT) and pretrained language models (PLM,e.g. T5). Most previous works use pretrained language models with subword-basedtokenization. However, the granularity of input units affects the alignment ofspeech model outputs and language model inputs, and PLM with character-basedtokenization is underexplored. In this work, we conduct extensive studies onhow PLMs with different tokenization strategies affect spoken languageunderstanding task including spoken question answering (SQA) and speechtranslation (ST). We further extend the idea to create T5lephone(pronounced astelephone), a variant of T5 that is pretrained using phonemicized text. Weinitialize T5lephone with existing PLMs to pretrain it using relativelylightweight computational resources. We reached state-of-the-art on NMSQA, andthe T5lephone model exceeds T5 with other types of units on end-to-end SQA andST.

T5lephone: Bridging Speech and Text Self-supervised Models for Spoken Language Understanding via Phoneme level T5