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

Reverberation negatively impacts the performance of automatic speechrecognition (ASR). Prior work on quantifying the effect of reverberation hasshown that clarity (C50), a parameter that can be estimated from the acousticimpulse response, is correlated with ASR performance. In this paper we proposepredicting ASR performance in terms of the word error rate (WER) directly fromacoustic parameters via a polynomial, sigmoidal, or neural network fit, as wellas blindly from reverberant speech samples using a convolutional neural network(CNN). We carry out experiments on two state-of-the-art ASR models and a largeset of acoustic impulse responses (AIRs). The results confirm C50 and C80 to behighly correlated with WER, allowing WER to be predicted with the proposedfitting approaches. The proposed non-intrusive CNN model outperforms C50-basedWER prediction, indicating that WER can be estimated blindly, i.e., directlyfrom the reverberant speech samples without knowledge of the acousticparameters.

Predicting word error rate for reverberant speech