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

In this paper, we propose an end-to-end post-filter method with deepattention fusion features for monaural speaker-independent speech separation.At first, a time-frequency domain speech separation method is applied as thepre-separation stage. The aim of pre-separation stage is to separate themixture preliminarily. Although this stage can separate the mixture, it stillcontains the residual interference. In order to enhance the pre-separatedspeech and improve the separation performance further, the end-to-endpost-filter (E2EPF) with deep attention fusion features is proposed. The E2EPFcan make full use of the prior knowledge of the pre-separated speech, whichcontributes to speech separation. It is a fully convolutional speech separationnetwork and uses the waveform as the input features. Firstly, the 1-Dconvolutional layer is utilized to extract the deep representation features forthe mixture and pre-separated signals in the time domain. Secondly, to pay moreattention to the outputs of the pre-separation stage, an attention module isapplied to acquire deep attention fusion features, which are extracted bycomputing the similarity between the mixture and the pre-separated speech.These deep attention fusion features are conducive to reduce the interferenceand enhance the pre-separated speech. Finally, these features are sent to thepost-filter to estimate each target signals. Experimental results on theWSJ0-2mix dataset show that the proposed method outperforms thestate-of-the-art speech separation method. Compared with the pre-separationmethod, our proposed method can acquire 64.1%, 60.2%, 25.6% and 7.5% relativeimprovements in scale-invariant source-to-noise ratio (SI-SNR), thesignal-to-distortion ratio (SDR), the perceptual evaluation of speech quality(PESQ) and the short-time objective intelligibility (STOI) measures,respectively.

Deep Attention Fusion Feature for Speech Separation with End-to-End Post-filter Method