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

Zero-shot voice conversion (VC) aims to transfer the source speaker timbre toarbitrary unseen target speaker timbre, while keeping the linguistic contentunchanged. Although the voice of generated speech can be controlled byproviding the speaker embedding of the target speaker, the speaker similaritystill lags behind the ground truth recordings. In this paper, we proposeSEF-VC, a speaker embedding free voice conversion model, which is designed tolearn and incorporate speaker timbre from reference speech via a powerfulposition-agnostic cross-attention mechanism, and then reconstruct waveform fromHuBERT semantic tokens in a non-autoregressive manner. The concise design ofSEF-VC enhances its training stability and voice conversion performance.Objective and subjective evaluations demonstrate the superiority of SEF-VC togenerate high-quality speech with better similarity to target reference thanstrong zero-shot VC baselines, even for very short reference speeches.

SEF-VC: Speaker Embedding Free Zero-Shot Voice Conversion with Cross Attention