Traditional studies on voice conversion (VC) have made progress with parallel
training data and known speakers. Good voice conversion quality is obtained by
exploring better alignment modules or expressive mapping functions. In this
study, we investigate zero-shot VC from a novel perspective of self-supervised
disentangled speech representation learning. Specifically, we achieve the
disentanglement by balancing the information flow between global speaker
representation and time-varying content representation in a sequential
variational autoencoder (VAE). A zero-shot voice conversion is performed by
feeding an arbitrary speaker embedding and content embeddings to the VAE
decoder. Besides that, an on-the-fly data augmentation training strategy is
applied to make the learned representation noise invariant. On TIMIT and VCTK
datasets, we achieve state-of-the-art performance on both objective evaluation,
i.e., speaker verification (SV) on speaker embedding and content embedding, and
subjective evaluation, i.e., voice naturalness and similarity, and remains to
be robust even with noisy source/target utterances.

本研究提出利用自监督分离式语音表示学习的新视角进行零样本语音转换，通过在序列变分自编码器中平衡全局说话人表示和时变内容表示之间的信息流实现分离，并应用实时数据增强培训策略以使所学表示不受噪声干扰。在 TIMIT 和 VCTK 数据集上表现出了优越性能，不仅在客观评估上表现良好，而且在主观评估方面具有鲁棒性，即使在源 / 目标语音帧存在噪声的情况下仍能保持不错的性能。