Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386875471> ?p ?o ?g. }
Showing items 1 to 61 of
61
with 100 items per page.
- W4386875471 abstract "Style voice conversion aims to transform the style of source speech to a desired style according to real-world application demands. However, the current style voice conversion approach relies on pre-defined labels or reference speech to control the conversion process, which leads to limitations in style diversity or falls short in terms of the intuitive and interpretability of style representation. In this study, we propose PromptVC, a novel style voice conversion approach that employs a latent diffusion model to generate a style vector driven by natural language prompts. Specifically, the style vector is extracted by a style encoder during training, and then the latent diffusion model is trained independently to sample the style vector from noise, with this process being conditioned on natural language prompts. To improve style expressiveness, we leverage HuBERT to extract discrete tokens and replace them with the K-Means center embedding to serve as the linguistic content, which minimizes residual style information. Additionally, we deduplicate the same discrete token and employ a differentiable duration predictor to re-predict the duration of each token, which can adapt the duration of the same linguistic content to different styles. The subjective and objective evaluation results demonstrate the effectiveness of our proposed system." @default.
- W4386875471 created "2023-09-20" @default.
- W4386875471 creator A5004857861 @default.
- W4386875471 creator A5008939026 @default.
- W4386875471 creator A5009348332 @default.
- W4386875471 creator A5013928267 @default.
- W4386875471 creator A5015560758 @default.
- W4386875471 creator A5033976028 @default.
- W4386875471 creator A5042083053 @default.
- W4386875471 creator A5052863980 @default.
- W4386875471 creator A5058766870 @default.
- W4386875471 creator A5081164682 @default.
- W4386875471 date "2023-09-17" @default.
- W4386875471 modified "2023-09-27" @default.
- W4386875471 title "PromptVC: Flexible Stylistic Voice Conversion in Latent Space Driven by Natural Language Prompts" @default.
- W4386875471 doi "https://doi.org/10.48550/arxiv.2309.09262" @default.
- W4386875471 hasPublicationYear "2023" @default.
- W4386875471 type Work @default.
- W4386875471 citedByCount "0" @default.
- W4386875471 crossrefType "posted-content" @default.
- W4386875471 hasAuthorship W4386875471A5004857861 @default.
- W4386875471 hasAuthorship W4386875471A5008939026 @default.
- W4386875471 hasAuthorship W4386875471A5009348332 @default.
- W4386875471 hasAuthorship W4386875471A5013928267 @default.
- W4386875471 hasAuthorship W4386875471A5015560758 @default.
- W4386875471 hasAuthorship W4386875471A5033976028 @default.
- W4386875471 hasAuthorship W4386875471A5042083053 @default.
- W4386875471 hasAuthorship W4386875471A5052863980 @default.
- W4386875471 hasAuthorship W4386875471A5058766870 @default.
- W4386875471 hasAuthorship W4386875471A5081164682 @default.
- W4386875471 hasBestOaLocation W43868754711 @default.
- W4386875471 hasConcept C154945302 @default.
- W4386875471 hasConcept C166957645 @default.
- W4386875471 hasConcept C204321447 @default.
- W4386875471 hasConcept C2776445246 @default.
- W4386875471 hasConcept C28490314 @default.
- W4386875471 hasConcept C41008148 @default.
- W4386875471 hasConcept C95457728 @default.
- W4386875471 hasConceptScore W4386875471C154945302 @default.
- W4386875471 hasConceptScore W4386875471C166957645 @default.
- W4386875471 hasConceptScore W4386875471C204321447 @default.
- W4386875471 hasConceptScore W4386875471C2776445246 @default.
- W4386875471 hasConceptScore W4386875471C28490314 @default.
- W4386875471 hasConceptScore W4386875471C41008148 @default.
- W4386875471 hasConceptScore W4386875471C95457728 @default.
- W4386875471 hasLocation W43868754711 @default.
- W4386875471 hasOpenAccess W4386875471 @default.
- W4386875471 hasPrimaryLocation W43868754711 @default.
- W4386875471 hasRelatedWork W1512718085 @default.
- W4386875471 hasRelatedWork W1569841287 @default.
- W4386875471 hasRelatedWork W2293457016 @default.
- W4386875471 hasRelatedWork W2351428524 @default.
- W4386875471 hasRelatedWork W2368779261 @default.
- W4386875471 hasRelatedWork W2369308426 @default.
- W4386875471 hasRelatedWork W2789919619 @default.
- W4386875471 hasRelatedWork W3169305685 @default.
- W4386875471 hasRelatedWork W1551406738 @default.
- W4386875471 hasRelatedWork W2610387714 @default.
- W4386875471 isParatext "false" @default.
- W4386875471 isRetracted "false" @default.
- W4386875471 workType "article" @default.