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

Recently normalizing flows have been gaining traction in text-to-speech (TTS)and voice conversion (VC) due to their state-of-the-art (SOTA) performance.Normalizing flows are unsupervised generative models. In this paper, weintroduce supervision to the training process of normalizing flows, without theneed for parallel data. We call this training paradigm AutoEncoder NormalizingFlow (AE-Flow). It adds a reconstruction loss forcing the model to useinformation from the conditioning to reconstruct an audio sample. Our goal isto understand the impact of each component and find the right combination ofthe negative log-likelihood (NLL) and the reconstruction loss in trainingnormalizing flows with coupling blocks. For that reason we will compareflow-based mapping model trained with: (i) NLL loss, (ii) NLL andreconstruction losses, as well as (iii) reconstruction loss only. Additionally,we compare our model with SOTA VC baseline. The models are evaluated in termsof naturalness, speaker similarity, intelligibility in many-to-many andmany-to-any VC settings. The results show that the proposed training paradigmsystematically improves speaker similarity and naturalness when compared toregular training methods of normalizing flows. Furthermore, we show that ourmethod improves speaker similarity and intelligibility over thestate-of-the-art.

AE-Flow: AutoEncoder Normalizing Flow