Instrumental Assessment of Prosodic Quality for Text-to-Speech Signals

Author

Norrenbrock, Christoph R. ; Hinterleitner, Florian ; Heute, Ulrich ; Möller, Sebastian

Author_Institution

Digital Signal Process. & Syst. Theor. Group (DSS), Univ. of Kiel, Kiel, Germany

Volume

19

Issue

5

fYear

2012

fDate

5/1/2012 12:00:00 AM

Firstpage

255

Lastpage

258

Abstract

Formal parameters of speech prosody are investigated concerning their ability to estimate the perceptual quality of text-to-speech (TTS) signals. The study is carried out for the German language using a broad databasis comprising a wide range of TTS systems and text materials. 18 purely acoustic markers, derived from F_o and vocalic/consonantal durations, are analysed individually and in conjunction via cross-validated regression models. The F_o slope within voiced segments proves particularly useful when integrated in a nonlinear fashion, whereas measures of durational variation perform comparably weak. The results highlight a strong potential for instrumental estimation techniques of TTS quality.

Keywords

acoustic signal processing; estimation theory; natural language processing; regression analysis; speech synthesis; German language; TTS signals; TTS systems; acoustic markers; broad databases; consonantal durations; cross-validated regression models; durational variation; formal parameters; instrumental assessment; instrumental estimation techniques; nonlinear fashion; perceptual quality estimation; prosodic quality; speech prosody; text materials; text-to-speech signals; vocalic durations; voiced segments; Correlation; Databases; Instruments; Materials; Rhythm; Speech; Timing; Instrumental quality assessment; prosody; speech quality; text-to-speech (TTS);

fLanguage

English

Journal_Title

Signal Processing Letters, IEEE

Publisher

ieee

ISSN

1070-9908

Type

jour

DOI

10.1109/LSP.2012.2189562

Filename

6161607