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

BERT-based re-ranking and dense retrieval (DR) systems have been shown toimprove search effectiveness for spoken content retrieval (SCR). However, bothmethods can still show a reduction in effectiveness when using ASR transcriptsin comparison to accurate manual transcripts. We find that a known-item searchtask on the How2 dataset of spoken instruction videos shows a reduction in meanreciprocal rank (MRR) scores of 10-14%. As a potential method to reduce thisdisparity, we investigate the use of semi-supervised ASR transcripts and N-bestASR transcripts to mitigate ASR errors for spoken search using BERT-basedranking. Semi-supervised ASR transcripts brought 2-5.5% MRR improvements overstandard ASR transcripts and our N-best early fusion methods for BERT DRsystems improved MRR by 3-4%. Combining semi-supervised transcripts with N-bestearly fusion for BERT DR reduced the MRR gap in search effectiveness betweenmanual and ASR transcripts by more than 50% from 14.32% to 6.58%.

Improving Noise Robustness for Spoken Content Retrieval using Semi-supervised ASR and N-best Transcripts for BERT-based Ranking Models