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

Achieving high accuracy with low latency has always been a challenge instreaming end-to-end automatic speech recognition (ASR) systems. By attendingto more future contexts, a streaming ASR model achieves higher accuracy butresults in larger latency, which hurts the streaming performance. In theMask-CTC framework, an encoder network is trained to learn the featurerepresentation that anticipates long-term contexts, which is desirable forstreaming ASR. Mask-CTC-based encoder pre-training has been shown beneficial inachieving low latency and high accuracy for triggered attention-based ASR.However, the effectiveness of this method has not been demonstrated for variousmodel architectures, nor has it been verified that the encoder has the expectedlook-ahead capability to reduce latency. This study, therefore, examines theeffectiveness of Mask-CTCbased pre-training for models with differentarchitectures, such as Transformer-Transducer and contextual block streamingASR. We also discuss the effect of the proposed pre-training method onobtaining accurate output spike timing.

Mask-CTC-based Encoder Pre-training for Streaming End-to-End Speech Recognition