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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

This paper investigates the effectiveness of factorial speech processingmodels in noise-robust automatic speech recognition tasks. For this purpose,the paper proposes an idealistic approach for modeling state-conditionalobservation distribution of factorial models based on weighted stereo samples.This approach is an extension to previous single pass retraining for idealmodel compensation which is extended here to support multiple audio sources.Non-stationary noises can be considered as one of these audio sources withmultiple states. Experiments of this paper over the set A of the Aurora 2dataset show that recognition performance can be improved by thisconsideration. The improvement is significant in low signal to noise energyconditions, up to 4% absolute word recognition accuracy. In addition to thepower of the proposed method in accurate representation of state-conditionalobservation distribution, it has an important advantage over previous methodsby providing the opportunity to independently select feature spaces for bothsource and corrupted features. This opens a new window for seeking betterfeature spaces appropriate for noisy speech, independent from clean speechfeatures.

Modeling State-Conditional Observation Distribution using Weighted Stereo Samples for Factorial Speech Processing Models