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

Speech dereverberation is often an important requirement in robust speechprocessing tasks. Supervised deep learning (DL) models give state-of-the-artperformance for single-channel speech dereverberation. Temporal convolutionalnetworks (TCNs) are commonly used for sequence modelling in speech enhancementtasks. A feature of TCNs is that they have a receptive field (RF) dependent onthe specific model configuration which determines the number of input framesthat can be observed to produce an individual output frame. It has been shownthat TCNs are capable of performing dereverberation of simulated speech data,however a thorough analysis, especially with focus on the RF is yet lacking inthe literature. This paper analyses dereverberation performance depending onthe model size and the RF of TCNs. Experiments using the WHAMR corpus which isextended to include room impulse responses (RIRs) with larger T60 valuesdemonstrate that a larger RF can have significant improvement in performancewhen training smaller TCN models. It is also demonstrated that TCNs benefitfrom a wider RF when dereverberating RIRs with larger RT60 values.

Receptive Field Analysis of Temporal Convolutional Networks for Monaural Speech Dereverberation