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

Generating versatile and appropriate synthetic speech requires control overthe output expression separate from the spoken text. Important non-textualspeech variation is seldom annotated, in which case output control must belearned in an unsupervised fashion. In this paper, we perform an in-depth studyof methods for unsupervised learning of control in statistical speechsynthesis. For example, we show that popular unsupervised training heuristicscan be interpreted as variational inference in certain autoencoder models. Weadditionally connect these models to VQ-VAEs, another, recently-proposed classof deep variational autoencoders, which we show can be derived from a verysimilar mathematical argument. The implications of these new probabilisticinterpretations are discussed. We illustrate the utility of the variousapproaches with an application to acoustic modelling for emotional speechsynthesis, where the unsupervised methods for learning expression control(without access to emotional labels) are found to give results that in manyaspects match or surpass the previous best supervised approach.

Deep Encoder-Decoder Models for Unsupervised Learning of Controllable Speech Synthesis