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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

We consider the problem of speaker diarization, the problem of segmenting anaudio recording of a meeting into temporal segments corresponding to individualspeakers. The problem is rendered particularly difficult by the fact that weare not allowed to assume knowledge of the number of people participating inthe meeting. To address this problem, we take a Bayesian nonparametric approachto speaker diarization that builds on the hierarchical Dirichlet process hiddenMarkov model (HDP-HMM) of Teh et al. [J. Amer. Statist. Assoc. 101 (2006)1566--1581]. Although the basic HDP-HMM tends to over-segment the audiodata---creating redundant states and rapidly switching among them---we describean augmented HDP-HMM that provides effective control over the switching rate.We also show that this augmentation makes it possible to treat emissiondistributions nonparametrically. To scale the resulting architecture torealistic diarization problems, we develop a sampling algorithm that employs atruncated approximation of the Dirichlet process to jointly resample the fullstate sequence, greatly improving mixing rates. Working with a benchmark NISTdata set, we show that our Bayesian nonparametric architecture yieldsstate-of-the-art speaker diarization results.

A sticky HDP-HMM with application to speaker diarization