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

Conventional speech enhancement technique such as beamforming has knownbenefits for far-field speech recognition. Our own work in frequency-domainmulti-channel acoustic modeling has shown additional improvements by training aspatial filtering layer jointly within an acoustic model. In this paper, wefurther develop this idea and use frequency aligned network for robustmulti-channel automatic speech recognition (ASR). Unlike an affine layer in thefrequency domain, the proposed frequency aligned component prevents onefrequency bin influencing other frequency bins. We show that this modificationnot only reduces the number of parameters in the model but also significantlyand improves the ASR performance. We investigate effects of frequency alignednetwork through ASR experiments on the real-world far-field data where usersare interacting with an ASR system in uncontrolled acoustic environments. Weshow that our multi-channel acoustic model with a frequency aligned networkshows up to 18% relative reduction in word error rate.

Robust Multi-channel Speech Recognition using Frequency Aligned Network