Synthesis of the intonation of neutrally spoken Modern Standard Arabic speech

Authors:
Yousif A. El-Imam
Affiliations:
Faculty of Engineering, Department of Electrical and Computer Engineering, The University of Sharjah, Sharjah, P. O. Box 27272, United Arab Emirates
Venue:
Signal Processing
Year:
2008

Citing 3
Cited 1

Practical approaches to speech coding

Practical approaches to speech coding
Prosodic modelling in Swedish speech synthesis

Speech Communication - Special issue: Fujisaki's Festschrift
Classification of methods used for assessment of text-to-speech systems according to the demands placed on the listener

Speech Communication

Automatic diacritization for low-resource languages using a hybrid word and consonant CMM

HLT '10 Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics

Quantified Score

Hi-index	0.08

Visualization

Abstract

Acoustical analyses of the fundamental frequency (F"0) contours of neutrally spoken Modern Standard Arabic (MSA) speech types of declarative, imperative, exclamative, and interrogative nature showed that their pitch patterns are characterized by four attributes: fluctuations around a mean pitch value that lies along either a declining or a constant line, a narrowing dynamic pitch range, an isochrony period between successive accented syllables, and a two-phase pitch rise in case of certain interrogatives. As such the intonation patterns of Arabic speech are mathematically modeled and specified by parameters, which were obtained from natural F"0 contours. Using the intonation model, continuous synthesized intonation curves were generated that resemble natural intonation patterns. Intonation synthesis has applications in speech synthesis as it can be used to give natural flavor to artificially produced speech. Thus the synthesized intonation curves were imposed on synthesized speech of the types considered, and subjective perceptual tests using mean opinion scores (MOS) revealed that synthesized speech produced with synthesized intonation is perceptually of comparable naturalness to synthesized speech produced with natural intonation.