Evaluation of Tacotron Based Synthesizers for Spanish and Basque
dc.contributor.author | García Romillo, Víctor | |
dc.contributor.author | Hernáez Rioja, Inmaculada ![]() | |
dc.contributor.author | Navas Cordón, Eva ![]() | |
dc.date.accessioned | 2022-02-11T17:30:55Z | |
dc.date.available | 2022-02-11T17:30:55Z | |
dc.date.issued | 2022-02-07 | |
dc.identifier.citation | Applied Sciences 12 (3): 1686 (2022) | es_ES |
dc.identifier.issn | 2076-3417 | |
dc.identifier.uri | http://hdl.handle.net/10810/55452 | |
dc.description.abstract | In this paper, we describe the implementation and evaluation of Text to Speech synthesizers based on neural networks for Spanish and Basque. Several voices were built, all of them using a limited number of data. The system applies Tacotron 2 to compute mel-spectrograms from the input sequence, followed by WaveGlow as neural vocoder to obtain the audio signals from the spectrograms. The limited number of data used for training the models leads to synthesis errors in some sentences. To automatically detect those errors, we developed a new method that is able to find the sentences that have lost the alignment during the inference process. To mitigate the problem, we implemented a guided attention providing the system with the explicit duration of the phonemes. The resulting system was evaluated to assess its robustness, quality and naturalness both with objective and subjective measures. The results reveal the capacity of the system to produce good quality and natural audios. | es_ES |
dc.description.sponsorship | This work was funded by the Basque Government (Project refs. PIBA 2018-035, IT-1355-19). This work is part of the project Grant PID 2019-108040RB-C21 funded by MCIN/AEI/10.13039/ 501100011033. | es_ES |
dc.language.iso | eng | es_ES |
dc.publisher | MDPI | es_ES |
dc.relation | info:eu-repo/grantAgreement/MCIN/PID 2019-108040RB-C21 | es_ES |
dc.rights | info:eu-repo/semantics/openAccess | es_ES |
dc.rights.uri | http://creativecommons.org/licenses/by/3.0/es/ | |
dc.subject | speech synthesis | es_ES |
dc.subject | robustness | es_ES |
dc.subject | text to speech | es_ES |
dc.subject | Spanish | es_ES |
dc.subject | Basque | es_ES |
dc.title | Evaluation of Tacotron Based Synthesizers for Spanish and Basque | es_ES |
dc.type | info:eu-repo/semantics/article | es_ES |
dc.date.updated | 2022-02-11T14:46:20Z | |
dc.rights.holder | © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). | es_ES |
dc.relation.publisherversion | https://www.mdpi.com/2076-3417/12/3/1686 | es_ES |
dc.identifier.doi | 10.3390/app12031686 | |
dc.departamentoes | Ingeniería de comunicaciones | |
dc.departamentoeu | Ingeniaritza kimikoa |
Files in this item
This item appears in the following Collection(s)
Except where otherwise noted, this item's license is described as © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).