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

The design of acoustic features is important for speech separation. It can beroughly categorized into three classes: handcrafted, parameterized, andlearnable features. Among them, learnable features, which are trained withseparation networks jointly in an end-to-end fashion, become a new trend ofmodern speech separation research, e.g. convolutional time domain audioseparation network (Conv-Tasnet), while handcrafted and parameterized featuresare also shown competitive in very recent studies. However, a systematiccomparison across the three kinds of acoustic features has not been conductedyet. In this paper, we compare them in the framework of Conv-Tasnet by settingits encoder and decoder with different acoustic features. We also generalizethe handcrafted multi-phase gammatone filterbank (MPGTF) to a new parameterizedmulti-phase gammatone filterbank (ParaMPGTF). Experimental results on theWSJ0-2mix corpus show that (i) if the decoder is learnable, then setting theencoder to STFT, MPGTF, ParaMPGTF, and learnable features lead to similarperformance; and (ii) when the pseudo-inverse transforms of STFT, MPGTF, andParaMPGTF are used as the decoders, the proposed ParaMPGTF performs better thanthe other two handcrafted features.

A comparison of handcrafted, parameterized, and learnable features for speech separation