{"ID":2878503,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.17868","arxiv_id":"2508.17868","title":"FasterVoiceGrad: Faster One-step Diffusion-Based Voice Conversion with Adversarial Diffusion Conversion Distillation","abstract":"A diffusion-based voice conversion (VC) model (e.g., VoiceGrad) can achieve high speech quality and speaker similarity; however, its conversion process is slow owing to iterative sampling. FastVoiceGrad overcomes this limitation by distilling VoiceGrad into a one-step diffusion model. However, it still requires a computationally intensive content encoder to disentangle the speaker's identity and content, which slows conversion. Therefore, we propose FasterVoiceGrad, a novel one-step diffusion-based VC model obtained by simultaneously distilling a diffusion model and content encoder using adversarial diffusion conversion distillation (ADCD), where distillation is performed in the conversion process while leveraging adversarial and score distillation training. Experimental evaluations of one-shot VC demonstrated that FasterVoiceGrad achieves competitive VC performance compared to FastVoiceGrad, with 6.6-6.9 and 1.8 times faster speed on a GPU and CPU, respectively.","short_abstract":"A diffusion-based voice conversion (VC) model (e.g., VoiceGrad) can achieve high speech quality and speaker similarity; however, its conversion process is slow owing to iterative sampling. FastVoiceGrad overcomes this limitation by distilling VoiceGrad into a one-step diffusion model. However, it still requires a compu...","url_abs":"https://arxiv.org/abs/2508.17868","url_pdf":"https://arxiv.org/pdf/2508.17868v1","authors":"[\"Takuhiro Kaneko\",\"Hirokazu Kameoka\",\"Kou Tanaka\",\"Yuto Kondo\"]","published":"2025-08-25T10:23:24Z","proceeding":"cs.SD","tasks":"[\"cs.SD\",\"cs.AI\",\"cs.LG\",\"eess.AS\",\"stat.ML\"]","methods":"[\"Diffusion Model\"]","has_code":false}