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

Previous multilingual text-to-speech (TTS) approaches have consideredleveraging monolingual speaker data to enable cross-lingual speech synthesis.However, such data-efficient approaches have ignored synthesizing emotionalaspects of speech due to the challenges of cross-speaker cross-lingual emotiontransfer - the heavy entanglement of speaker timbre, emotion, and languagefactors in the speech signal will make a system produce cross-lingual syntheticspeech with an undesired foreign accent and weak emotion expressiveness. Thispaper proposes the Multilingual Emotional TTS (METTS) model to mitigate theseproblems, realizing both cross-speaker and cross-lingual emotion transfer.Specifically, METTS takes DelightfulTTS as the backbone model and proposes thefollowing designs. First, to alleviate the foreign accent problem, METTSintroduces multi-scale emotion modeling to disentangle speech prosody intocoarse-grained and fine-grained scales, producing language-agnostic andlanguage-specific emotion representations, respectively. Second, as apre-processing step, formant shift-based information perturbation is applied tothe reference signal for better disentanglement of speaker timbre in thespeech. Third, a vector quantization-based emotion matcher is designed forreference selection, leading to decent naturalness and emotion diversity incross-lingual synthetic speech. Experiments demonstrate the good design ofMETTS.

METTS: Multilingual Emotional Text-to-Speech by Cross-speaker and Cross-lingual Emotion Transfer