Matches in SemOpenAlex for { <https://semopenalex.org/work/W4298017177> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W4298017177 abstract "In this paper, we propose VISinger, a complete end-to-end high-quality singing voice synthesis (SVS) system that directly generates audio waveform from lyrics and musical score. Our approach is inspired by VITS, which adopts VAE-based posterior encoder augmented with normalizing flow-based prior encoder and adversarial decoder to realize complete end-to-end speech generation. VISinger follows the main architecture of VITS, but makes substantial improvements to the prior encoder based on the characteristics of singing. First, instead of using phoneme-level mean and variance of acoustic features, we introduce a length regulator and a frame prior network to get the frame-level mean and variance on acoustic features, modeling the rich acoustic variation in singing. Second, we further introduce an F0 predictor to guide the frame prior network, leading to stabler singing performance. Finally, to improve the singing rhythm, we modify the duration predictor to specifically predict the phoneme to note duration ratio, helped with singing note normalization. Experiments on a professional Mandarin singing corpus show that VISinger significantly outperforms FastSpeech+Neural-Vocoder two-stage approach and the oracle VITS; ablation study demonstrates the effectiveness of different contributions." @default.
- W4298017177 created "2022-10-01" @default.
- W4298017177 creator A5029817092 @default.
- W4298017177 creator A5036369578 @default.
- W4298017177 creator A5047688996 @default.
- W4298017177 creator A5050166453 @default.
- W4298017177 creator A5060467975 @default.
- W4298017177 creator A5064426059 @default.
- W4298017177 date "2021-10-17" @default.
- W4298017177 modified "2023-09-27" @default.
- W4298017177 title "VISinger: Variational Inference with Adversarial Learning for End-to-End Singing Voice Synthesis" @default.
- W4298017177 doi "https://doi.org/10.48550/arxiv.2110.08813" @default.
- W4298017177 hasPublicationYear "2021" @default.
- W4298017177 type Work @default.
- W4298017177 citedByCount "0" @default.
- W4298017177 crossrefType "posted-content" @default.
- W4298017177 hasAuthorship W4298017177A5029817092 @default.
- W4298017177 hasAuthorship W4298017177A5036369578 @default.
- W4298017177 hasAuthorship W4298017177A5047688996 @default.
- W4298017177 hasAuthorship W4298017177A5050166453 @default.
- W4298017177 hasAuthorship W4298017177A5060467975 @default.
- W4298017177 hasAuthorship W4298017177A5064426059 @default.
- W4298017177 hasBestOaLocation W42980171771 @default.
- W4298017177 hasConcept C112758219 @default.
- W4298017177 hasConcept C115903868 @default.
- W4298017177 hasConcept C121332964 @default.
- W4298017177 hasConcept C126042441 @default.
- W4298017177 hasConcept C136886441 @default.
- W4298017177 hasConcept C144024400 @default.
- W4298017177 hasConcept C154945302 @default.
- W4298017177 hasConcept C19165224 @default.
- W4298017177 hasConcept C24890656 @default.
- W4298017177 hasConcept C2779803651 @default.
- W4298017177 hasConcept C28490314 @default.
- W4298017177 hasConcept C41008148 @default.
- W4298017177 hasConcept C44819458 @default.
- W4298017177 hasConcept C50644808 @default.
- W4298017177 hasConcept C55166926 @default.
- W4298017177 hasConcept C76155785 @default.
- W4298017177 hasConcept C94915269 @default.
- W4298017177 hasConceptScore W4298017177C112758219 @default.
- W4298017177 hasConceptScore W4298017177C115903868 @default.
- W4298017177 hasConceptScore W4298017177C121332964 @default.
- W4298017177 hasConceptScore W4298017177C126042441 @default.
- W4298017177 hasConceptScore W4298017177C136886441 @default.
- W4298017177 hasConceptScore W4298017177C144024400 @default.
- W4298017177 hasConceptScore W4298017177C154945302 @default.
- W4298017177 hasConceptScore W4298017177C19165224 @default.
- W4298017177 hasConceptScore W4298017177C24890656 @default.
- W4298017177 hasConceptScore W4298017177C2779803651 @default.
- W4298017177 hasConceptScore W4298017177C28490314 @default.
- W4298017177 hasConceptScore W4298017177C41008148 @default.
- W4298017177 hasConceptScore W4298017177C44819458 @default.
- W4298017177 hasConceptScore W4298017177C50644808 @default.
- W4298017177 hasConceptScore W4298017177C55166926 @default.
- W4298017177 hasConceptScore W4298017177C76155785 @default.
- W4298017177 hasConceptScore W4298017177C94915269 @default.
- W4298017177 hasLocation W42980171771 @default.
- W4298017177 hasOpenAccess W4298017177 @default.
- W4298017177 hasPrimaryLocation W42980171771 @default.
- W4298017177 hasRelatedWork W1833279920 @default.
- W4298017177 hasRelatedWork W2067342401 @default.
- W4298017177 hasRelatedWork W2201063935 @default.
- W4298017177 hasRelatedWork W2736042296 @default.
- W4298017177 hasRelatedWork W2894174195 @default.
- W4298017177 hasRelatedWork W3012498027 @default.
- W4298017177 hasRelatedWork W3035430139 @default.
- W4298017177 hasRelatedWork W3110041771 @default.
- W4298017177 hasRelatedWork W3200345197 @default.
- W4298017177 hasRelatedWork W3207340675 @default.
- W4298017177 isParatext "false" @default.
- W4298017177 isRetracted "false" @default.
- W4298017177 workType "article" @default.