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

Speaker diarization has been investigated extensively as an important centraltask for meeting analysis. Recent trend shows that integration of end-to-endneural (EEND)-and clustering-based diarization is a promising approach tohandle realistic conversational data containing overlapped speech with anarbitrarily large number of speakers, and achieved state-of-the-art results onvarious tasks. However, the approaches proposed so far have not realized {\ittight} integration yet, because the clustering employed therein was not optimalin any sense for clustering the speaker embeddings estimated by the EENDmodule. To address this problem, this paper introduces a {\it trainable}clustering algorithm into the integration framework, by deep-unfolding anon-parametric Bayesian model called the infinite Gaussian mixture model(iGMM). Specifically, the speaker embeddings are optimized during training suchthat it better fits iGMM clustering, based on a novel clustering loss based onAdjusted Rand Index (ARI). Experimental results based on CALLHOME data showthat the proposed approach outperforms the conventional approach in terms ofdiarization error rate (DER), especially by substantially reducing speakerconfusion errors, that indeed reflects the effectiveness of the proposed iGMMintegration.

Tight integration of neural- and clustering-based diarization through deep unfolding of infinite Gaussian mixture model