Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385571418> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W4385571418 abstract "Generative adversarial networks (GANs) and denoising diffusion probabilistic models (DDPMs) have recently achieved impressive performances in image and audio synthesis. After revisiting their success in conditional speech synthesis, we find that 1) GANs sacrifice sample diversity for quality and speed, 2) diffusion models exhibit outperformed sample quality and diversity at a high computational cost, where achieving high-quality, fast, and diverse speech synthesis challenges all neural synthesizers. In this work, we propose to converge advantages from GANs and diffusion models by incorporating both classes, introducing dual-empowered modeling perspectives: 1) FastDiff 2 (DiffGAN), a diffusion model whose denoising process is parametrized by conditional GANs, and the non-Gaussian denoising distribution makes it much more stable to implement the reverse process with large steps sizes; and 2) FastDiff 2 (GANDiff), a generative adversarial network whose forward process is constructed by multiple denoising diffusion iterations, which exhibits better sample diversity than traditional GANs. Experimental results show that both variants enjoy an efficient 4-step sampling process and demonstrate superior sample quality and diversity. Audio samples are available at https://RevisitSpeech.github.io/" @default.
- W4385571418 created "2023-08-05" @default.
- W4385571418 creator A5002111548 @default.
- W4385571418 creator A5012657658 @default.
- W4385571418 creator A5059358267 @default.
- W4385571418 creator A5065126806 @default.
- W4385571418 creator A5065361552 @default.
- W4385571418 creator A5083611896 @default.
- W4385571418 date "2023-01-01" @default.
- W4385571418 modified "2023-10-17" @default.
- W4385571418 title "FastDiff 2: Revisiting and Incorporating GANs and Diffusion Models in High-Fidelity Speech Synthesis" @default.
- W4385571418 doi "https://doi.org/10.18653/v1/2023.findings-acl.437" @default.
- W4385571418 hasPublicationYear "2023" @default.
- W4385571418 type Work @default.
- W4385571418 citedByCount "0" @default.
- W4385571418 crossrefType "proceedings-article" @default.
- W4385571418 hasAuthorship W4385571418A5002111548 @default.
- W4385571418 hasAuthorship W4385571418A5012657658 @default.
- W4385571418 hasAuthorship W4385571418A5059358267 @default.
- W4385571418 hasAuthorship W4385571418A5065126806 @default.
- W4385571418 hasAuthorship W4385571418A5065361552 @default.
- W4385571418 hasAuthorship W4385571418A5083611896 @default.
- W4385571418 hasBestOaLocation W43855714181 @default.
- W4385571418 hasConcept C111919701 @default.
- W4385571418 hasConcept C113364801 @default.
- W4385571418 hasConcept C11413529 @default.
- W4385571418 hasConcept C119599485 @default.
- W4385571418 hasConcept C119857082 @default.
- W4385571418 hasConcept C121332964 @default.
- W4385571418 hasConcept C127413603 @default.
- W4385571418 hasConcept C154945302 @default.
- W4385571418 hasConcept C163294075 @default.
- W4385571418 hasConcept C185592680 @default.
- W4385571418 hasConcept C198531522 @default.
- W4385571418 hasConcept C23224414 @default.
- W4385571418 hasConcept C2776459999 @default.
- W4385571418 hasConcept C39890363 @default.
- W4385571418 hasConcept C41008148 @default.
- W4385571418 hasConcept C43617362 @default.
- W4385571418 hasConcept C49937458 @default.
- W4385571418 hasConcept C69357855 @default.
- W4385571418 hasConcept C76155785 @default.
- W4385571418 hasConcept C97355855 @default.
- W4385571418 hasConcept C98045186 @default.
- W4385571418 hasConceptScore W4385571418C111919701 @default.
- W4385571418 hasConceptScore W4385571418C113364801 @default.
- W4385571418 hasConceptScore W4385571418C11413529 @default.
- W4385571418 hasConceptScore W4385571418C119599485 @default.
- W4385571418 hasConceptScore W4385571418C119857082 @default.
- W4385571418 hasConceptScore W4385571418C121332964 @default.
- W4385571418 hasConceptScore W4385571418C127413603 @default.
- W4385571418 hasConceptScore W4385571418C154945302 @default.
- W4385571418 hasConceptScore W4385571418C163294075 @default.
- W4385571418 hasConceptScore W4385571418C185592680 @default.
- W4385571418 hasConceptScore W4385571418C198531522 @default.
- W4385571418 hasConceptScore W4385571418C23224414 @default.
- W4385571418 hasConceptScore W4385571418C2776459999 @default.
- W4385571418 hasConceptScore W4385571418C39890363 @default.
- W4385571418 hasConceptScore W4385571418C41008148 @default.
- W4385571418 hasConceptScore W4385571418C43617362 @default.
- W4385571418 hasConceptScore W4385571418C49937458 @default.
- W4385571418 hasConceptScore W4385571418C69357855 @default.
- W4385571418 hasConceptScore W4385571418C76155785 @default.
- W4385571418 hasConceptScore W4385571418C97355855 @default.
- W4385571418 hasConceptScore W4385571418C98045186 @default.
- W4385571418 hasLocation W43855714181 @default.
- W4385571418 hasOpenAccess W4385571418 @default.
- W4385571418 hasPrimaryLocation W43855714181 @default.
- W4385571418 hasRelatedWork W2017572488 @default.
- W4385571418 hasRelatedWork W2053559412 @default.
- W4385571418 hasRelatedWork W2378174816 @default.
- W4385571418 hasRelatedWork W2468514837 @default.
- W4385571418 hasRelatedWork W2920684403 @default.
- W4385571418 hasRelatedWork W3108083835 @default.
- W4385571418 hasRelatedWork W4225288502 @default.
- W4385571418 hasRelatedWork W4361806674 @default.
- W4385571418 hasRelatedWork W4365600489 @default.
- W4385571418 hasRelatedWork W4385242011 @default.
- W4385571418 isParatext "false" @default.
- W4385571418 isRetracted "false" @default.
- W4385571418 workType "article" @default.