Exportar Publicação

Daniela Braga, Pedro, S., Ribeiro, M., Henriques, M. & Dias, J. (2008). HMM-based Brazilian Portuguese TTS. In Daniela Braga, Miguel Sales Dias, António Teixeira (Ed.), Proc. Propor 2008 Special Session  (Computational Processing of the Portuguese Language)- Applications of Portuguese Speech and Language  Technologies . (pp. 48-51). Curia: Microsoft Language Development Center.

D. Braga et al.,  "HMM-based Brazilian Portuguese TTS", in Proc. Propor 2008 Special Session  (Computational Processing of the Portuguese Language)- Applications of Portuguese Speech and Language  Technologies , Daniela Braga, Miguel Sales Dias, António Teixeira, Ed., Curia, Microsoft Language Development Center, 2008, pp. 48-51

@inproceedings{braga2008_1732210793478,
	author = "Daniela Braga and Pedro, S. and Ribeiro, M. and Henriques, M. and Dias, J.",
	title = "HMM-based Brazilian Portuguese TTS",
	booktitle = "Proc. Propor 2008 Special Session  (Computational Processing of the Portuguese Language)- Applications of Portuguese Speech and Language  Technologies ",
	year = "2008",
	editor = "Daniela Braga, Miguel Sales Dias, António Teixeira",
	volume = "",
	number = "",
	series = "",
	pages = "48-51",
	publisher = "Microsoft Language Development Center",
	address = "Curia",
	organization = "",
	url = "https://link.springer.com/book/10.1007/978-3-540-85980-2"
}

TY  - CPAPER
TI  - HMM-based Brazilian Portuguese TTS
T2  - Proc. Propor 2008 Special Session  (Computational Processing of the Portuguese Language)- Applications of Portuguese Speech and Language  Technologies 
AU  - Daniela Braga
AU  - Pedro, S.
AU  - Ribeiro, M.
AU  - Henriques, M.
AU  - Dias, J.
PY  - 2008
SP  - 48-51
CY  - Curia
UR  - https://link.springer.com/book/10.1007/978-3-540-85980-2
AB  - In this paper, a Hidden Markov Models-based Text-to-Speech system (hereafter HTS) for Brazilian Portuguese (BP) was 
presented. The Brazilian Portuguese Text-to-Speech system (hereafter BP TTS), within other languages, will be enabled in 
several Microsoft products whose release date is still being planning. The architecture of the system was described as well as 
the main issues arising from this specific language development. The test plan and procedures were presented and its results 
were discussed. The comprehension tests carried out by 7 listeners gave rise to 98.915% of overall intelligibility rate 
distributed by the following domains: e-mail/sentences, addresses, person’s names, date and time and news/paragraphs. 
Another comprehension test was carried out with a competitor commercial system and the obtained overall intelligibility rate 
was 91.95%.
ER  -