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

The recently proposed serialized output training (SOT) simplifiesmulti-talker automatic speech recognition (ASR) by generating speakertranscriptions separated by a special token. However, frequent speaker changescan make speaker change prediction difficult. To address this, we proposeboundary-aware serialized output training (BA-SOT), which explicitlyincorporates boundary knowledge into the decoder via a speaker change detectiontask and boundary constraint loss. We also introduce a two-stage connectionisttemporal classification (CTC) strategy that incorporates token-level SOT CTC torestore temporal context information. Besides typical character error rate(CER), we introduce utterance-dependent character error rate (UD-CER) tofurther measure the precision of speaker change prediction. Compared tooriginal SOT, BA-SOT reduces CER/UD-CER by 5.1%/14.0%, and leveraging apre-trained ASR model for BA-SOT model initialization further reducesCER/UD-CER by 8.4%/19.9%.

BA-SOT: Boundary-Aware Serialized Output Training for Multi-Talker ASR