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

Since its inception, the field of deep speech enhancement has been dominatedby predictive (discriminative) approaches, such as spectral mapping or masking.Recently, however, novel generative approaches have been applied to speechenhancement, attaining good denoising performance with high subjective qualityscores. At the same time, advances in deep learning also allowed for thecreation of neural network-based metrics, which have desirable traits such asbeing able to work without a reference (non-intrusively). Since generativelyenhanced speech tends to exhibit radically different residual distortions, itsevaluation using instrumental speech metrics may behave differently compared topredictively enhanced speech. In this paper, we evaluate the performance of thesame speech enhancement backbone trained under predictive and generativeparadigms on a variety of metrics and show that intrusive and non-intrusivemeasures correlate differently for each paradigm. This analysis motivates thesearch for metrics that can together paint a complete and unbiased picture ofspeech enhancement performance, irrespective of the model's training process.

On the Behavior of Intrusive and Non-intrusive Speech Enhancement Metrics in Predictive and Generative Settings