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

Connectionist Temporal Classification (CTC) suffers from the latency problemwhen applied to streaming models. We argue that in CTC lattice, the alignmentsthat can access more future context are preferred during training, therebyleading to higher symbol delay. In this work we propose the delay-penalized CTCwhich is augmented with latency penalty regularization. We devise a flexibleand efficient implementation based on the differentiable Finite StateTransducer (FST). Specifically, by attaching a binary attribute to CTCtopology, we can locate the frames that firstly emit non-blank tokens on theresulting CTC lattice, and add the frame offsets to the log-probabilities.Experimental results demonstrate the effectiveness of our proposeddelay-penalized CTC, which is able to balance the delay-accuracy trade-off.Furthermore, combining the delay-penalized transducer enables the CTC model toachieve better performance and lower latency. Our work is open-sourced andpublicly available https://github.com/k2-fsa/k2.

Delay-penalized CTC implemented based on Finite State Transducer