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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

This paper describes a new unsupervised machine learning method forsimultaneous phoneme and word discovery from multiple speakers. Human infantscan acquire knowledge of phonemes and words from interactions with his/hermother as well as with others surrounding him/her. From a computationalperspective, phoneme and word discovery from multiple speakers is a morechallenging problem than that from one speaker because the speech signals fromdifferent speakers exhibit different acoustic features. This paper proposes anunsupervised phoneme and word discovery method that simultaneously usesnonparametric Bayesian double articulation analyzer (NPB-DAA) and deep sparseautoencoder with parametric bias in hidden layer (DSAE-PBHL). We assume that aninfant can recognize and distinguish speakers based on certain other features,e.g., visual face recognition. DSAE-PBHL is aimed to be able to subtractspeaker-dependent acoustic features and extract speaker-independent features.An experiment demonstrated that DSAE-PBHL can subtract distributedrepresentations of acoustic signals, enabling extraction based on the types ofphonemes rather than on the speakers. Another experiment demonstrated that acombination of NPB-DAA and DSAE-PB outperformed the available methods inphoneme and word discovery tasks involving speech signals with Japanese vowelsequences from multiple speakers.

Unsupervised Phoneme and Word Discovery from Multiple Speakers using Double Articulation Analyzer and Neural Network with Parametric Bias