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

This paper proposes a novel label-synchronous speech-to-text alignmenttechnique for automatic speech recognition (ASR). The speech-to-text alignmentis a problem of splitting long audio recordings with un-aligned transcriptsinto utterance-wise pairs of speech and text. Unlike conventional methods basedon frame-synchronous prediction, the proposed method re-defines thespeech-to-text alignment as a label-synchronous text mapping problem. Thisenables an accurate alignment benefiting from the strong inference ability ofthe state-of-the-art attention-based encoder-decoder models, which cannot beapplied to the conventional methods. Two different Transformer models namedforward Transformer and backward Transformer are respectively used forestimating an initial and final tokens of a given speech segment based onend-of-sentence prediction with teacher-forcing. Experiments using the corpusof spontaneous Japanese (CSJ) demonstrate that the proposed method provides anaccurate utterance-wise alignment, that matches the manually annotatedalignment with as few as 0.2% errors. It is also confirmed that aTransformer-based hybrid CTC/Attention ASR model using the aligned speech andtext pairs as an additional training data reduces character error ratesrelatively up to 59.0%, which is significantly better than 39.0% reduction by aconventional alignment method based on connectionist temporal classificationmodel.

Label-Synchronous Speech-to-Text Alignment for ASR Using Forward and Backward Transformers