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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

In this work, we focus on improving ASR output segmentation in the context oflow-resource language speech-to-text translation. ASR output segmentation iscrucial, as ASR systems segment the input audio using purely acousticinformation and are not guaranteed to output sentence-like segments. Since mostMT systems expect sentences as input, feeding in longer unsegmented passagescan lead to sub-optimal performance. We explore the feasibility of usingdatasets of subtitles from TV shows and movies to train better ASR segmentationmodels. We further incorporate part-of-speech (POS) tag and dependency labelinformation (derived from the unsegmented ASR outputs) into our segmentationmodel. We show that this noisy syntactic information can improve modelaccuracy. We evaluate our models intrinsically on segmentation quality andextrinsically on downstream MT performance, as well as downstream tasksincluding cross-lingual information retrieval (CLIR) tasks and human relevanceassessments. Our model shows improved performance on downstream tasks forLithuanian and Bulgarian.

Subtitles to Segmentation: Improving Low-Resource Speech-to-Text Translation Pipelines