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

While the performance of cross-lingual TTS based on monolingual corpora hasbeen significantly improved recently, generating cross-lingual speech stillsuffers from the foreign accent problem, leading to limited naturalness.Besides, current cross-lingual methods ignore modeling emotion, which isindispensable paralinguistic information in speech delivery. In this paper, wepropose DiCLET-TTS, a Diffusion model based Cross-Lingual Emotion Transfermethod that can transfer emotion from a source speaker to the intra- andcross-lingual target speakers. Specifically, to relieve the foreign accentproblem while improving the emotion expressiveness, the terminal distributionof the forward diffusion process is parameterized into a speaker-irrelevant butemotion-related linguistic prior by a prior text encoder with the emotionembedding as a condition. To address the weaker emotional expressivenessproblem caused by speaker disentanglement in emotion embedding, a novelorthogonal projection based emotion disentangling module (OP-EDM) is proposedto learn the speaker-irrelevant but emotion-discriminative embedding. Moreover,a condition-enhanced DPM decoder is introduced to strengthen the modelingability of the speaker and the emotion in the reverse diffusion process tofurther improve emotion expressiveness in speech delivery. Cross-lingualemotion transfer experiments show the superiority of DiCLET-TTS over variouscompetitive models and the good design of OP-EDM in learning speaker-irrelevantbut emotion-discriminative embedding.

DiCLET-TTS: Diffusion Model based Cross-lingual Emotion Transfer for Text-to-Speech -- A Study between English and Mandarin