Fastvoicegrad: One-step Diffusion-based Voice Conversion With Adversarial Conditional Diffusion Distillation
2024 · Takuhiro Kaneko, Hirokazu Kameoka, Kou Tanaka, et al.
Abstract
Diffusion-based voice conversion (VC) techniques such as VoiceGrad have attracted interest because of their high VC performance in terms of speech quality and speaker similarity. However, a notable limitation is the slow inference caused by the multi-step reverse diffusion. Therefore, we propose FastVoiceGrad, a novel one-step diffusion-based VC that reduces the number of iterations from dozens to one while inheriting the high VC performance of the multi-step diffusion-based VC. We obtain the model using adversarial conditional diffusion distillation (ACDD), leveraging the ability of generative adversarial networks and diffusion models while reconsidering the initial states in sampling. Evaluations of one-shot any-to-any VC demonstrate that FastVoiceGrad achieves VC performance superior to or comparable to that of previous multi-step diffusion-based VC while enhancing the inference speed. Audio samples are available at https://www.kecl.ntt.co.jp/people/kaneko.takuhiro/projects/fastvoic
Authors
(none)
Tags
Stats
Related papers
- Voicegrad: Non-parallel Any-to-many Voice Conversion With Annealed Langevin Dynamics (2020)7.50
- Diffusion-based Voice Conversion With Fast Maximum Likelihood Sampling Scheme (2021)0.00
- Fastvc: Fast Voice Conversion With Non-parallel Data (2020)5.24
- Diff-hiervc: Diffusion-based Hierarchical Voice Conversion With Robust Pitch Generation And Masked Prior For Zero-shot Speaker Adaptation (2023)0.00
- Stepback: Enhanced Disentanglement For Voice Conversion Via Multi-task Learning (2025)0.00
- Codiff-vc: A Codec-assisted Diffusion Model For Zero-shot Voice Conversion (2024)0.00
- Converting Anyone's Voice: End-to-end Expressive Voice Conversion With A Conditional Diffusion Model (2024)5.24
- An Adaptive Learning Based Generative Adversarial Network For One-to-one Voice Conversion (2021)10.61