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/

Zendy Plus

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

Zendy Tools

Zendy Open

This paper aims to build a multi-speaker expressive TTS system, synthesizinga target speaker's speech with multiple styles and emotions. To this end, wepropose a novel contrastive learning-based TTS approach to transfer style andemotion across speakers. Specifically, contrastive learning from differentlevels, i.e. utterance and category level, is leveraged to extract thedisentangled style, emotion, and speaker representations from speech for styleand emotion transfer. Furthermore, a semi-supervised training strategy isintroduced to improve the data utilization efficiency by involving multi-domaindata, including style-labeled data, emotion-labeled data, and abundantunlabeled data. To achieve expressive speech with diverse styles and emotionsfor a target speaker, the learned disentangled representations are integratedinto an improved VITS model. Experiments on multi-domain data demonstrate theeffectiveness of the proposed method.

Boosting Multi-Speaker Expressive Speech Synthesis with Semi-supervised Contrastive Learning