Matches in SemOpenAlex for { <https://semopenalex.org/work/W3154451338> ?p ?o ?g. }
- W3154451338 endingPage "1302" @default.
- W3154451338 startingPage "1290" @default.
- W3154451338 abstract "We present a novel voice conversion (VC) framework by learning from a text-to-speech (TTS) synthesis system, that is called TTS-VC transfer learning or TTL-VC for short. We first develop a multi-speaker speech synthesis system with sequence-to-sequence encoder-decoder architecture, where the encoder extracts the linguistic representations of input text, while the decoder, conditioned on target speaker embedding, takes the context vectors and the attention recurrent network cell output to generate target acoustic features. We take advantage of the fact that TTS system maps input text to speaker independent context vectors, thus re-purpose such a mapping to supervise the training of the latent representations of an encoder-decoder voice conversion system. In the voice conversion system, the encoder takes speech instead of text as the input, while the decoder is functionally similar to the TTS decoder. As we condition the decoder on a speaker embedding, the system can be trained on non-parallel data for any-to-any voice conversion. During voice conversion training, we present both text and speech to speech synthesis and voice conversion networks respectively. At run-time, the voice conversion network uses its own encoder-decoder architecture without the need of text input. Experiments show that the proposed TTL-VC system outperforms two competitive voice conversion baselines consistently, namely phonetic posteriorgram and AutoVC methods, in terms of speech quality, naturalness, and speaker similarity." @default.
- W3154451338 created "2021-04-26" @default.
- W3154451338 creator A5032690182 @default.
- W3154451338 creator A5065562660 @default.
- W3154451338 creator A5069292664 @default.
- W3154451338 creator A5070281927 @default.
- W3154451338 date "2021-01-01" @default.
- W3154451338 modified "2023-10-16" @default.
- W3154451338 title "Transfer Learning From Speech Synthesis to Voice Conversion With Non-Parallel Training Data" @default.
- W3154451338 cites W1509691205 @default.
- W3154451338 cites W1517202054 @default.
- W3154451338 cites W1977362459 @default.
- W3154451338 cites W1991402032 @default.
- W3154451338 cites W2013996527 @default.
- W3154451338 cites W2022125261 @default.
- W3154451338 cites W2046056978 @default.
- W3154451338 cites W2056852181 @default.
- W3154451338 cites W2057609679 @default.
- W3154451338 cites W2076055233 @default.
- W3154451338 cites W2086796102 @default.
- W3154451338 cites W2102003408 @default.
- W3154451338 cites W2107860279 @default.
- W3154451338 cites W2120605154 @default.
- W3154451338 cites W2123003832 @default.
- W3154451338 cites W2123808477 @default.
- W3154451338 cites W2150769028 @default.
- W3154451338 cites W2150933458 @default.
- W3154451338 cites W2151262064 @default.
- W3154451338 cites W2161135987 @default.
- W3154451338 cites W2161476805 @default.
- W3154451338 cites W2165108269 @default.
- W3154451338 cites W2169652224 @default.
- W3154451338 cites W2171019095 @default.
- W3154451338 cites W2471520273 @default.
- W3154451338 cites W2518172956 @default.
- W3154451338 cites W2532494225 @default.
- W3154451338 cites W2733416080 @default.
- W3154451338 cites W2749262635 @default.
- W3154451338 cites W2804998325 @default.
- W3154451338 cites W2806000759 @default.
- W3154451338 cites W2889064624 @default.
- W3154451338 cites W2901254300 @default.
- W3154451338 cites W2902070858 @default.
- W3154451338 cites W2937579788 @default.
- W3154451338 cites W2941094131 @default.
- W3154451338 cites W2962788625 @default.
- W3154451338 cites W2963539064 @default.
- W3154451338 cites W2963609956 @default.
- W3154451338 cites W2963830550 @default.
- W3154451338 cites W2964069186 @default.
- W3154451338 cites W2964243274 @default.
- W3154451338 cites W2972359262 @default.
- W3154451338 cites W2972689158 @default.
- W3154451338 cites W2972999331 @default.
- W3154451338 cites W3006777338 @default.
- W3154451338 cites W3015805741 @default.
- W3154451338 cites W3015826515 @default.
- W3154451338 cites W3034089333 @default.
- W3154451338 cites W3095936335 @default.
- W3154451338 cites W3095990227 @default.
- W3154451338 cites W3096567388 @default.
- W3154451338 cites W3096864844 @default.
- W3154451338 cites W3099078140 @default.
- W3154451338 cites W3101689408 @default.
- W3154451338 cites W3125709657 @default.
- W3154451338 doi "https://doi.org/10.1109/taslp.2021.3066047" @default.
- W3154451338 hasPublicationYear "2021" @default.
- W3154451338 type Work @default.
- W3154451338 sameAs 3154451338 @default.
- W3154451338 citedByCount "22" @default.
- W3154451338 countsByYear W31544513382021 @default.
- W3154451338 countsByYear W31544513382022 @default.
- W3154451338 countsByYear W31544513382023 @default.
- W3154451338 crossrefType "journal-article" @default.
- W3154451338 hasAuthorship W3154451338A5032690182 @default.
- W3154451338 hasAuthorship W3154451338A5065562660 @default.
- W3154451338 hasAuthorship W3154451338A5069292664 @default.
- W3154451338 hasAuthorship W3154451338A5070281927 @default.
- W3154451338 hasBestOaLocation W31544513381 @default.
- W3154451338 hasConcept C111919701 @default.
- W3154451338 hasConcept C118505674 @default.
- W3154451338 hasConcept C121332964 @default.
- W3154451338 hasConcept C134537474 @default.
- W3154451338 hasConcept C14999030 @default.
- W3154451338 hasConcept C151730666 @default.
- W3154451338 hasConcept C154945302 @default.
- W3154451338 hasConcept C2779343474 @default.
- W3154451338 hasConcept C28490314 @default.
- W3154451338 hasConcept C41008148 @default.
- W3154451338 hasConcept C41608201 @default.
- W3154451338 hasConcept C62520636 @default.
- W3154451338 hasConcept C86803240 @default.
- W3154451338 hasConceptScore W3154451338C111919701 @default.
- W3154451338 hasConceptScore W3154451338C118505674 @default.
- W3154451338 hasConceptScore W3154451338C121332964 @default.
- W3154451338 hasConceptScore W3154451338C134537474 @default.
- W3154451338 hasConceptScore W3154451338C14999030 @default.
- W3154451338 hasConceptScore W3154451338C151730666 @default.