DocumentCode
2139916
Title
Speech synthesis module for Romanian language
Author
Gavat, Inge ; Zirra, Matei ; Grigore, Mihaela
Author_Institution
Politech. Univ. of Bucharest, Romania
Volume
2
fYear
1997
fDate
7-11 Oct 1997
Firstpage
593
Abstract
The architecture of a speech synthesis module based on the formant model of speech production is proposed. The module can be entirely controlled by a low stream of coded parameters, from less than 1.5 Kb/s to produce synthetic speech, with good intelligibility and acceptable naturalness. The model can be used in voice response systems, in voiced alarms, for multimedia purposes and represents a possibility to extend speech from the man to man communication also to the machine-man communication
Keywords
speech coding; speech intelligibility; speech synthesis; vocoders; 1.5 Kbit/s; Romanian language; acceptable naturalness; formant model; frame bit allocation; intelligibility; low stream of coded parameters; machine-man communication; multimedia; speech production; speech synthesis module; synthetic speech; text-to-speech; voice response systems; voiced alarms; Bandwidth; Frequency; Humans; Natural languages; Speech analysis; Speech coding; Speech enhancement; Speech synthesis; Streaming media; Synthesizers;
fLanguage
English
Publisher
ieee
Conference_Titel
Semiconductor Conference, 1997. CAS '97 Proceedings., 1997 International
Conference_Location
Sinaia
Print_ISBN
0-7803-3804-9
Type
conf
DOI
10.1109/SMICND.1997.651331
Filename
651331
Link To Document