FastVoiceGrad: One-step Diffusion-Based Voice Conversion with Adversarial Conditional Diffusion Distillation
This addresses a bottleneck in voice conversion for applications requiring real-time processing, though it is incremental as it builds on existing diffusion techniques.
The paper tackles the slow inference of diffusion-based voice conversion by proposing FastVoiceGrad, a one-step method that reduces iterations from dozens to one while maintaining or improving performance in speech quality and speaker similarity.
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/fastvoicegrad/.