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 work we aim to discover high quality speech features and linguisticunits directly from unlabeled speech data in a zero resource scenario. Theresults are evaluated using the metrics and corpora proposed in the ZeroResource Speech Challenge organized at Interspeech 2015. A Multi-layeredAcoustic Tokenizer (MAT) was proposed for automatic discovery of multiple setsof acoustic tokens from the given corpus. Each acoustic token set is specifiedby a set of hyperparameters that describe the model configuration. These setsof acoustic tokens carry different characteristics fof the given corpus and thelanguage behind, thus can be mutually reinforced. The multiple sets of tokenlabels are then used as the targets of a Multi-target Deep Neural Network(MDNN) trained on low-level acoustic features. Bottleneck features extractedfrom the MDNN are then used as the feedback input to the MAT and the MDNNitself in the next iteration. We call this iterative deep learning frameworkthe Multi-layered Acoustic Tokenizing Deep Neural Network (MAT-DNN), whichgenerates both high quality speech features for the Track 1 of the Challengeand acoustic tokens for the Track 2 of the Challenge. In addition, we performedextra experiments on the same corpora on the application of query-by-examplespoken term detection. The experimental results showed the iterative deeplearning framework of MAT-DNN improved the detection performance due to betterunderlying speech features and acoustic tokens.

An Iterative Deep Learning Framework for Unsupervised Discovery of Speech Features and Linguistic Units with Applications on Spoken Term Detection