Matches in SemOpenAlex for { <https://semopenalex.org/work/W2239904444> ?p ?o ?g. }
- W2239904444 abstract "Recently, recurrent neural networks (RNNs) as powerful sequence models have re-emerged as a potential acoustic model for statistical parametric speech synthesis (SPSS). The long short-term memory (LSTM) architecture is particularly attractive because it addresses the vanishing gradient problem in standard RNNs, making them easier to train. Although recent studies have demonstrated that LSTMs can achieve significantly better performance on SPSS than deep feed-forward neural networks, little is known about why. Here we attempt to answer two questions: a) why do LSTMs work well as a sequence model for SPSS; b) which component (e.g., input gate, output gate, forget gate) is most important. We present a visual analysis alongside a series of experiments, resulting in a proposal for a simplified architecture. The simplified architecture has significantly fewer parameters than an LSTM, thus reducing generation complexity considerably without degrading quality." @default.
- W2239904444 created "2016-06-24" @default.
- W2239904444 creator A5009522499 @default.
- W2239904444 creator A5062516688 @default.
- W2239904444 date "2016-01-11" @default.
- W2239904444 modified "2023-09-27" @default.
- W2239904444 title "Investigating gated recurrent neural networks for speech synthesis" @default.
- W2239904444 cites W1499332833 @default.
- W2239904444 cites W1543299179 @default.
- W2239904444 cites W1571950845 @default.
- W2239904444 cites W1576227399 @default.
- W2239904444 cites W1689711448 @default.
- W2239904444 cites W1924770834 @default.
- W2239904444 cites W1990505856 @default.
- W2239904444 cites W2045158511 @default.
- W2239904444 cites W2049686551 @default.
- W2239904444 cites W2102003408 @default.
- W2239904444 cites W2106564373 @default.
- W2239904444 cites W2129142580 @default.
- W2239904444 cites W2134973740 @default.
- W2239904444 cites W2150658333 @default.
- W2239904444 cites W2294797155 @default.
- W2239904444 cites W2395700867 @default.
- W2239904444 cites W2397670047 @default.
- W2239904444 cites W2399057275 @default.
- W2239904444 cites W2949190276 @default.
- W2239904444 cites W2950635152 @default.
- W2239904444 cites W581956982 @default.
- W2239904444 hasPublicationYear "2016" @default.
- W2239904444 type Work @default.
- W2239904444 sameAs 2239904444 @default.
- W2239904444 citedByCount "4" @default.
- W2239904444 countsByYear W22399044442016 @default.
- W2239904444 countsByYear W22399044442018 @default.
- W2239904444 crossrefType "posted-content" @default.
- W2239904444 hasAuthorship W2239904444A5009522499 @default.
- W2239904444 hasAuthorship W2239904444A5062516688 @default.
- W2239904444 hasConcept C105795698 @default.
- W2239904444 hasConcept C117251300 @default.
- W2239904444 hasConcept C121332964 @default.
- W2239904444 hasConcept C123657996 @default.
- W2239904444 hasConcept C133488467 @default.
- W2239904444 hasConcept C142362112 @default.
- W2239904444 hasConcept C147168706 @default.
- W2239904444 hasConcept C14999030 @default.
- W2239904444 hasConcept C153349607 @default.
- W2239904444 hasConcept C154945302 @default.
- W2239904444 hasConcept C168167062 @default.
- W2239904444 hasConcept C24574437 @default.
- W2239904444 hasConcept C2778112365 @default.
- W2239904444 hasConcept C28490314 @default.
- W2239904444 hasConcept C33923547 @default.
- W2239904444 hasConcept C41008148 @default.
- W2239904444 hasConcept C50644808 @default.
- W2239904444 hasConcept C54355233 @default.
- W2239904444 hasConcept C86803240 @default.
- W2239904444 hasConcept C97355855 @default.
- W2239904444 hasConceptScore W2239904444C105795698 @default.
- W2239904444 hasConceptScore W2239904444C117251300 @default.
- W2239904444 hasConceptScore W2239904444C121332964 @default.
- W2239904444 hasConceptScore W2239904444C123657996 @default.
- W2239904444 hasConceptScore W2239904444C133488467 @default.
- W2239904444 hasConceptScore W2239904444C142362112 @default.
- W2239904444 hasConceptScore W2239904444C147168706 @default.
- W2239904444 hasConceptScore W2239904444C14999030 @default.
- W2239904444 hasConceptScore W2239904444C153349607 @default.
- W2239904444 hasConceptScore W2239904444C154945302 @default.
- W2239904444 hasConceptScore W2239904444C168167062 @default.
- W2239904444 hasConceptScore W2239904444C24574437 @default.
- W2239904444 hasConceptScore W2239904444C2778112365 @default.
- W2239904444 hasConceptScore W2239904444C28490314 @default.
- W2239904444 hasConceptScore W2239904444C33923547 @default.
- W2239904444 hasConceptScore W2239904444C41008148 @default.
- W2239904444 hasConceptScore W2239904444C50644808 @default.
- W2239904444 hasConceptScore W2239904444C54355233 @default.
- W2239904444 hasConceptScore W2239904444C86803240 @default.
- W2239904444 hasConceptScore W2239904444C97355855 @default.
- W2239904444 hasLocation W22399044441 @default.
- W2239904444 hasOpenAccess W2239904444 @default.
- W2239904444 hasPrimaryLocation W22399044441 @default.
- W2239904444 hasRelatedWork W1499864241 @default.
- W2239904444 hasRelatedWork W1576227399 @default.
- W2239904444 hasRelatedWork W1689711448 @default.
- W2239904444 hasRelatedWork W2064675550 @default.
- W2239904444 hasRelatedWork W2284692871 @default.
- W2239904444 hasRelatedWork W2402268235 @default.
- W2239904444 hasRelatedWork W2524611247 @default.
- W2239904444 hasRelatedWork W2559959364 @default.
- W2239904444 hasRelatedWork W2566275344 @default.
- W2239904444 hasRelatedWork W2613006358 @default.
- W2239904444 hasRelatedWork W2749432513 @default.
- W2239904444 hasRelatedWork W2883775970 @default.
- W2239904444 hasRelatedWork W2901056065 @default.
- W2239904444 hasRelatedWork W2950689855 @default.
- W2239904444 hasRelatedWork W2963770546 @default.
- W2239904444 hasRelatedWork W2964060510 @default.
- W2239904444 hasRelatedWork W2982408358 @default.
- W2239904444 hasRelatedWork W3014997656 @default.
- W2239904444 hasRelatedWork W3015357480 @default.
- W2239904444 hasRelatedWork W581956982 @default.