Matches in SemOpenAlex for { <https://semopenalex.org/work/W1783473872> ?p ?o ?g. }
Showing items 1 to 85 of
85
with 100 items per page.
- W1783473872 abstract "This paper describes the use of a low-dimensional vector representation of sentence acoustics to control the output of a feed-forward deep neural network text-to-speech system on a sentence-by-sentence basis. Vector representations for sentences in the training corpus are learned during network training along with other parameters of the model. Although the network is trained on a frame-by-frame basis, the standard framelevel inputs representing linguistic features are supplemented by features from a projection layer which outputs a learned representation of sentence-level acoustic characteristics. The projection layer contains dedicated parameters for each sentence in the training data which are optimised jointly with the standard network weights. Sentence-specific parameters are optimised on all frames of the relevant sentence – these parameters therefore allow the network to account for sentence-level variation in the data which is not predictable from the standard linguistic inputs. Results show that the global prosodic characteristics of synthetic speech can be controlled simply and robustly at run time by supplementing basic linguistic features with sentencelevel control vectors which are novel but designed to be consistent with those observed in the training corpus." @default.
- W1783473872 created "2016-06-24" @default.
- W1783473872 creator A5009522499 @default.
- W1783473872 creator A5049904155 @default.
- W1783473872 creator A5062516688 @default.
- W1783473872 date "2015-09-06" @default.
- W1783473872 modified "2023-10-18" @default.
- W1783473872 title "Sentence-level control vectors for deep neural network speech synthesis" @default.
- W1783473872 cites W141956881 @default.
- W1783473872 cites W1499332833 @default.
- W1783473872 cites W1854240507 @default.
- W1783473872 cites W1970689298 @default.
- W1783473872 cites W1977362129 @default.
- W1783473872 cites W1993409002 @default.
- W1783473872 cites W1996430422 @default.
- W1783473872 cites W2015687745 @default.
- W1783473872 cites W2039800941 @default.
- W1783473872 cites W2042600255 @default.
- W1783473872 cites W2074854222 @default.
- W1783473872 cites W2090755665 @default.
- W1783473872 cites W2102003408 @default.
- W1783473872 cites W2112021726 @default.
- W1783473872 cites W2116082100 @default.
- W1783473872 cites W2123442489 @default.
- W1783473872 cites W2132339004 @default.
- W1783473872 cites W2134973740 @default.
- W1783473872 cites W2145247325 @default.
- W1783473872 cites W2154920538 @default.
- W1783473872 cites W2156692643 @default.
- W1783473872 cites W2296111744 @default.
- W1783473872 cites W2396043161 @default.
- W1783473872 cites W80543058 @default.
- W1783473872 cites W1971670143 @default.
- W1783473872 doi "https://doi.org/10.21437/interspeech.2015-128" @default.
- W1783473872 hasPublicationYear "2015" @default.
- W1783473872 type Work @default.
- W1783473872 sameAs 1783473872 @default.
- W1783473872 citedByCount "29" @default.
- W1783473872 countsByYear W17834738722016 @default.
- W1783473872 countsByYear W17834738722017 @default.
- W1783473872 countsByYear W17834738722018 @default.
- W1783473872 countsByYear W17834738722019 @default.
- W1783473872 countsByYear W17834738722020 @default.
- W1783473872 countsByYear W17834738722021 @default.
- W1783473872 countsByYear W17834738722022 @default.
- W1783473872 countsByYear W17834738722023 @default.
- W1783473872 crossrefType "proceedings-article" @default.
- W1783473872 hasAuthorship W1783473872A5009522499 @default.
- W1783473872 hasAuthorship W1783473872A5049904155 @default.
- W1783473872 hasAuthorship W1783473872A5062516688 @default.
- W1783473872 hasBestOaLocation W17834738722 @default.
- W1783473872 hasConcept C14999030 @default.
- W1783473872 hasConcept C154945302 @default.
- W1783473872 hasConcept C204321447 @default.
- W1783473872 hasConcept C2775924081 @default.
- W1783473872 hasConcept C2777530160 @default.
- W1783473872 hasConcept C28490314 @default.
- W1783473872 hasConcept C41008148 @default.
- W1783473872 hasConcept C50644808 @default.
- W1783473872 hasConceptScore W1783473872C14999030 @default.
- W1783473872 hasConceptScore W1783473872C154945302 @default.
- W1783473872 hasConceptScore W1783473872C204321447 @default.
- W1783473872 hasConceptScore W1783473872C2775924081 @default.
- W1783473872 hasConceptScore W1783473872C2777530160 @default.
- W1783473872 hasConceptScore W1783473872C28490314 @default.
- W1783473872 hasConceptScore W1783473872C41008148 @default.
- W1783473872 hasConceptScore W1783473872C50644808 @default.
- W1783473872 hasLocation W17834738721 @default.
- W1783473872 hasLocation W17834738722 @default.
- W1783473872 hasOpenAccess W1783473872 @default.
- W1783473872 hasPrimaryLocation W17834738721 @default.
- W1783473872 hasRelatedWork W2009831055 @default.
- W1783473872 hasRelatedWork W2083530853 @default.
- W1783473872 hasRelatedWork W2146114872 @default.
- W1783473872 hasRelatedWork W2368686738 @default.
- W1783473872 hasRelatedWork W2375873920 @default.
- W1783473872 hasRelatedWork W2389383932 @default.
- W1783473872 hasRelatedWork W2392060890 @default.
- W1783473872 hasRelatedWork W2392760275 @default.
- W1783473872 hasRelatedWork W2393172683 @default.
- W1783473872 hasRelatedWork W2982905616 @default.
- W1783473872 isParatext "false" @default.
- W1783473872 isRetracted "false" @default.
- W1783473872 magId "1783473872" @default.
- W1783473872 workType "article" @default.