Title: GMM classification of text-to-speech synthesis: identification of original speaker’s voice
Other Titles: Klasifikace syntĂ©zy Ĺ™eÄŤi z textu pomocĂ­ GMM: Identifikace pĹŻvodnĂ­ho hlasu Ĺ™eÄŤnĂ­ka
Authors: Přibil, Jiří
Přibilová, Anna
Matoušek, Jindřich
Citation: PĹIBIL, Jiří; PĹIBILOVĂ, Anna; MATOUĹ EK, JindĹ™ich. GMM classification of text-to-speech synthesis: identification of original speaker’s voice. In: Text, speech and dialogue. Berlin: Springer, 2014, p. 365-373. (Lecture notes in artificial intelligence; 8655). ISBN 978-3-319-10815-5.
Issue Date: 2014
Publisher: Springer
Document type: článek
URI: http://hdl.handle.net/11025/26010
ISBN: 978-3-319-10815-5
ISSN: 0302-9743
Keywords: syntĂ©za Ĺ™eÄŤi;GMM klasifikace;statistická analĂ˝za
Keywords in different language: speech synthesis;GMM classification;statistical analysis
Abstract in different language: This paper describes two experiments. The first one deals with evaluation of synthetic speech quality by reverse identification of original speakers whose voices had been used for several Czech text-to-speech (TTS) systems. The second experiment was aimed at evaluation of the influence of voice transformation on the original speaker recognition. The paper further describes an analysis of the influence of initial settings for creation and training of the Gaussian mixture models (GMM), and the influence of different types of used speech features (spectral and/or supra-segmental) on correctness of GMM identification. The stability of the identification process with respect to the duration of the tested sentence (number of the processed frames) was analysed, too.
Rights: Â© Springer
Appears in Collections:Články / Articles (KIV)

Files in This Item:
File Description SizeFormat 
Pribil.pdfPlný text569,11 kBAdobe PDFView/Open

Please use this identifier to cite or link to this item: http://hdl.handle.net/11025/26010

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.