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 investigate a novel cross-lingual multi-speaker text-to-speech synthesisapproach for generating high-quality native or accented speech fornative/foreign seen/unseen speakers in English and Mandarin. The systemconsists of three separately trained components: an x-vector speaker encoder, aTacotron-based synthesizer and a WaveNet vocoder. It is conditioned on 3 kindsof embeddings: (1) speaker embedding so that the system can be trained withspeech from many speakers will little data from each speaker; (2) languageembedding with shared phoneme inputs; (3) stress and tone embedding whichimproves naturalness of synthesized speech, especially for a tonal languagelike Mandarin. By adjusting the various embeddings, MOS results show that ourmethod can generate high-quality natural and intelligible native speech fornative/foreign seen/unseen speakers. Intelligibility and naturalness ofaccented speech is low as expected. Speaker similarity is good for nativespeech from native speakers. Interestingly, speaker similarity is also good foraccented speech from foreign speakers. We also find that normalizing speakerembedding x-vectors by L2-norm normalization or whitening improves outputquality a lot in many cases, and the WaveNet performance seems to belanguage-independent: our WaveNet is trained with Cantonese speech and can beused to generate Mandarin and English speech very well.

Cross-lingual Multi-speaker Text-to-speech Synthesis for Voice Cloning without Using Parallel Corpus for Unseen Speakers