Papers › AUTOVC: Zero-Shot Voice Style Transfer with Only Autoencoder Loss

AUTOVC: Zero-Shot Voice Style Transfer with Only Autoencoder Loss

14 May 2019arXiv:1905.05879archive 2025-07-28

Kaizhi Qian, Yang Zhang, Shiyu Chang, Xuesong Yang, Mark Hasegawa-Johnson

Non-parallel many-to-many voice conversion, as well as zero-shot voice conversion, remain under-explored areas. Deep style transfer algorithms, such as generative adversarial networks (GAN) and conditional variational autoencoder (CVAE), are being applied as new solutions in this field. However, GAN training is sophisticated and difficult, and there is no strong evidence that its generated speech is of good perceptual quality. On the other hand, CVAE training is simple but does not come with the distribution-matching property of a GAN. In this paper, we propose a new style transfer scheme that involves only an autoencoder with a carefully designed bottleneck. We formally show that this scheme can achieve distribution-matching style transfer by training only on a self-reconstruction loss. Based on this scheme, we proposed AUTOVC, which achieves state-of-the-art results in many-to-many voice conversion with non-parallel data, and which is the first to perform zero-shot voice conversion.

PaperPDFCode

Code

liusongxiang/StarGAN-Voice-Conversion officialmentioned in papermentioned on GitHubpytorch report
CODEJIN/AutoVC mentioned on GitHubpytorch report
RF5/simple-autovc mentioned on GitHubpytorch report
deciding/StarGAN-VC mentioned on GitHubpytorch report
freenowill/AutoVC-WavRNN mentioned on GitHubpytorch report
gkv856/end2end_auto_voice_conversion mentioned on GitHubpytorch report
sroutray/ugp mentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Style TransferVoice Conversion

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

ConvolutioncVAE

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections