A Comparison Of Discrete And Soft Speech Units For Improved Voice Conversion

Abstract

The goal of voice conversion is to transform source speech into a target voice, keeping the content unchanged. In this paper, we focus on self-supervised representation learning for voice conversion. Specifically, we compare discrete and soft speech units as input features. We find that discrete representations effectively remove speaker information but discard some linguistic content - leading to mispronunciations. As a solution, we propose soft speech units. To learn soft units, we predict a distribution over discrete speech units. By modeling uncertainty, soft units capture more content information, improving the intelligibility and naturalness of converted speech. Samples available at https://ubisoft-laforge.github.io/speech/soft-vc/. Code available at https://github.com/bshall/soft-vc/.

Stats

• citations97
• S2 citations—
• github stars456
• HF likes0
• heat score20.25
• arxiv keyvanniekerk2021a

Code

• bshall/soft-vc—

Related papers

• Unsupervised End-to-end Learning Of Discrete Linguistic Units For Voice Conversion (2019)9.03
• Discrete Unit Based Masking For Improving Disentanglement In Voice Conversion (2024)0.00
• ACE-VC: Adaptive And Controllable Voice Conversion Using Explicitly Disentangled Self-supervised Speech Representations (2023)0.00
• A Comparative Study Of Self-supervised Speech Representation Based Voice Conversion (2022)9.76
• Selfvc: Voice Conversion With Iterative Refinement Using Self Transformations (2023)0.00
• Zero-shot Voice Conversion Via Self-supervised Prosody Representation Learning (2021)6.34
• Robust Disentangled Variational Speech Representation Learning For Zero-shot Voice Conversion (2022)10.97
• Learning Disentangled Speech Representations With Contrastive Learning And Time-invariant Retrieval (2024)5.84