Mostra el registre d'ítem simple
TTS evaluation campaign with a common spanish database
dc.contributor.author | Sainz, Iñaki |
dc.contributor.author | Navas, Eva |
dc.contributor.author | Hernáez, Inma |
dc.contributor.author | Bonafonte Cávez, Antonio |
dc.contributor.author | Campillo, Francisco |
dc.contributor.other | Universitat Politècnica de Catalunya. Departament de Teoria del Senyal i Comunicacions |
dc.date.accessioned | 2010-10-11T07:58:56Z |
dc.date.available | 2010-10-11T07:58:56Z |
dc.date.created | 2010 |
dc.date.issued | 2010 |
dc.identifier.citation | Sainz, I. [et al.]. TTS evaluation campaign with a common spanish database. A: International Conference on Language Resources and Evaluation. "Seventh Int. Conf. on Language Resources and Evaluation (LREC)". Valleta: 2010, p. 2155-2160. |
dc.identifier.isbn | 2-9517408-6-7 |
dc.identifier.uri | http://hdl.handle.net/2117/9608 |
dc.description.abstract | This paper describes the first TTS evaluation campaign designed for Spanish. Seven research institutions took part in the evaluation campaign and developed a voice from a common speech database provided by the organisation. Each participating team had a period of seven weeks to generate a voice. Next, a set of sentences were released and each team had to synthesise them within a week period. Finally, some of the synthesised test audio files were subjectively evaluated via an online test according to the following criteria: similarity to the original voice, naturalness and intelligibility. Box-plots, Wilcoxon tests and WER have been generated in order to analyse the results. Two main conclusions can be drawn: On the one hand, there is considerable margin for improvement to reach the quality level of the natural voice. On the other hand, two systems get significantly better results than the rest: one is based on statistical parametric synthesis and the other one is a concatenative system that makes use of a sinusoidal model to modify both prosody and smooth spectral joints. Therefore, it seems that some kind of spectral control is needed when building voices with a medium size database for unrestricted domains. |
dc.format.extent | 6 p. |
dc.language.iso | eng |
dc.rights | Attribution-NonCommercial-NoDerivs 3.0 Spain |
dc.rights.uri | http://creativecommons.org/licenses/by-nc-nd/3.0/es/ |
dc.subject | Àrees temàtiques de la UPC::Enginyeria de la telecomunicació |
dc.subject.lcsh | Text-to-speech software |
dc.subject.lcsh | Signal theory (Telecommunication) |
dc.title | TTS evaluation campaign with a common spanish database |
dc.type | Conference report |
dc.subject.lemac | Senyal, Teoria del (Telecomunicació) |
dc.contributor.group | Universitat Politècnica de Catalunya. VEU - Grup de Tractament de la Parla |
dc.relation.publisherversion | http://www.lrec-conf.org/proceedings/lrec2010/pdf/456_Paper.pdf |
dc.rights.access | Open Access |
local.identifier.drac | 3265447 |
dc.description.version | Postprint (published version) |
local.citation.author | Sainz, I.; Navas, E.; Hernáez, I.; Bonafonte, A.; Campillo, F. |
local.citation.contributor | International Conference on Language Resources and Evaluation |
local.citation.pubplace | Valleta |
local.citation.publicationName | Seventh Int. Conf. on Language Resources and Evaluation (LREC) |
local.citation.startingPage | 2155 |
local.citation.endingPage | 2160 |