FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3048379951> ?p ?o ?g. }

• W3048379951 abstract "Recently, sequence-to-sequence models with attention have been successfully applied in Text-to-speech (TTS). These models can generate near-human speech with a large accurately-transcribed speech corpus. However, preparing such a large data-set is both expensive and laborious. To alleviate the problem of heavy data demand, we propose a novel unsupervised pre-training mechanism in this paper. Specifically, we first use Vector-quantization Variational-Autoencoder (VQ-VAE) to ex-tract the unsupervised linguistic units from large-scale, publicly found, and untranscribed speech. We then pre-train the sequence-to-sequence TTS model by using the<unsupervised linguistic units, audio>pairs. Finally, we fine-tune the model with a small amount of<text, audio>paired data from the target speaker. As a result, both objective and subjective evaluations show that our proposed method can synthesize more intelligible and natural speech with the same amount of paired training data. Besides, we extend our proposed method to the hypothesized low-resource languages and verify the effectiveness of the method using objective evaluation." @default.
• W3048379951 created "2020-08-18" @default.
• W3048379951 creator A5018364104 @default.
• W3048379951 creator A5032164035 @default.
• W3048379951 date "2020-08-11" @default.
• W3048379951 modified "2023-09-23" @default.
• W3048379951 title "Unsupervised Learning For Sequence-to-sequence Text-to-speech For Low-resource Languages" @default.
• W3048379951 cites W1578102511 @default.
• W3048379951 cites W1783473872 @default.
• W3048379951 cites W2055408826 @default.
• W3048379951 cites W2099062766 @default.
• W3048379951 cites W2107789863 @default.
• W3048379951 cites W2107860279 @default.
• W3048379951 cites W2152859600 @default.
• W3048379951 cites W2400517318 @default.
• W3048379951 cites W2527729766 @default.
• W3048379951 cites W2746132399 @default.
• W3048379951 cites W2749074593 @default.
• W3048379951 cites W2765486990 @default.
• W3048379951 cites W2889028433 @default.
• W3048379951 cites W2901997113 @default.
• W3048379951 cites W2903853691 @default.
• W3048379951 cites W2913512413 @default.
• W3048379951 cites W2960427821 @default.
• W3048379951 cites W2962739369 @default.
• W3048379951 cites W2963192573 @default.
• W3048379951 cites W2963432880 @default.
• W3048379951 cites W2963609956 @default.
• W3048379951 cites W2963618559 @default.
• W3048379951 cites W2963799213 @default.
• W3048379951 cites W2963912924 @default.
• W3048379951 cites W2963975282 @default.
• W3048379951 cites W2964243274 @default.
• W3048379951 cites W2972428935 @default.
• W3048379951 cites W2972817752 @default.
• W3048379951 cites W2972943112 @default.
• W3048379951 cites W2973034126 @default.
• W3048379951 cites W3016159759 @default.
• W3048379951 cites W3030437843 @default.
• W3048379951 cites W3125709657 @default.
• W3048379951 doi "https://doi.org/10.48550/arxiv.2008.04549" @default.
• W3048379951 hasPublicationYear "2020" @default.
• W3048379951 type Work @default.
• W3048379951 sameAs 3048379951 @default.
• W3048379951 citedByCount "2" @default.
• W3048379951 countsByYear W30483799512020 @default.
• W3048379951 crossrefType "posted-content" @default.
• W3048379951 hasAuthorship W3048379951A5018364104 @default.
• W3048379951 hasAuthorship W3048379951A5032164035 @default.
• W3048379951 hasBestOaLocation W30483799511 @default.
• W3048379951 hasConcept C101738243 @default.
• W3048379951 hasConcept C108583219 @default.
• W3048379951 hasConcept C111919701 @default.
• W3048379951 hasConcept C118505674 @default.
• W3048379951 hasConcept C14999030 @default.
• W3048379951 hasConcept C154945302 @default.
• W3048379951 hasConcept C162324750 @default.
• W3048379951 hasConcept C187736073 @default.
• W3048379951 hasConcept C199833920 @default.
• W3048379951 hasConcept C204321447 @default.
• W3048379951 hasConcept C2778112365 @default.
• W3048379951 hasConcept C2780451532 @default.
• W3048379951 hasConcept C28490314 @default.
• W3048379951 hasConcept C35639132 @default.
• W3048379951 hasConcept C41008148 @default.
• W3048379951 hasConcept C54355233 @default.
• W3048379951 hasConcept C8038995 @default.
• W3048379951 hasConcept C86803240 @default.
• W3048379951 hasConceptScore W3048379951C101738243 @default.
• W3048379951 hasConceptScore W3048379951C108583219 @default.
• W3048379951 hasConceptScore W3048379951C111919701 @default.
• W3048379951 hasConceptScore W3048379951C118505674 @default.
• W3048379951 hasConceptScore W3048379951C14999030 @default.
• W3048379951 hasConceptScore W3048379951C154945302 @default.
• W3048379951 hasConceptScore W3048379951C162324750 @default.
• W3048379951 hasConceptScore W3048379951C187736073 @default.
• W3048379951 hasConceptScore W3048379951C199833920 @default.
• W3048379951 hasConceptScore W3048379951C204321447 @default.
• W3048379951 hasConceptScore W3048379951C2778112365 @default.
• W3048379951 hasConceptScore W3048379951C2780451532 @default.
• W3048379951 hasConceptScore W3048379951C28490314 @default.
• W3048379951 hasConceptScore W3048379951C35639132 @default.
• W3048379951 hasConceptScore W3048379951C41008148 @default.
• W3048379951 hasConceptScore W3048379951C54355233 @default.
• W3048379951 hasConceptScore W3048379951C8038995 @default.
• W3048379951 hasConceptScore W3048379951C86803240 @default.
• W3048379951 hasLocation W30483799511 @default.
• W3048379951 hasOpenAccess W3048379951 @default.
• W3048379951 hasPrimaryLocation W30483799511 @default.
• W3048379951 hasRelatedWork W2169373125 @default.
• W3048379951 hasRelatedWork W2959758584 @default.
• W3048379951 hasRelatedWork W3048379951 @default.
• W3048379951 hasRelatedWork W3090528452 @default.
• W3048379951 hasRelatedWork W3093058615 @default.
• W3048379951 hasRelatedWork W3096303254 @default.
• W3048379951 hasRelatedWork W3107474891 @default.
• W3048379951 hasRelatedWork W3195005325 @default.
• W3048379951 hasRelatedWork W4288282363 @default.
• W3048379951 hasRelatedWork W3161695192 @default.
• W3048379951 isParatext "false" @default.

• next