Sixth International Conference on Spoken Language Processing
(ICSLP 2000)

Beijing, China
October 16-20, 2000

Implementation of a Text-to-Speech System for Farsi Language

Hamid Reza Abutalebi (1,3), Mahmood Bijankhan (2,3)

(1) Dept. of Electrical Eng., Amirkabir University of Technology, (2) Dept. of Linguistics, University of Tehran, (3) Research Center of Intelligent Signal Processing (RCISP) Tehran, Iran

In this research, a Text-To-Speech system for Farsi language has been implemented. The proposed synthesizer concatenates Farsi syllables in a TD-PSOLA manner. This paper is mainly concentrated on investigation about pitch variations in Farsi sentences and presentation of some novel rules for modeling these variations. Based on the location of stressed syllable, we obtain a primary pitch curve for each word. Using prosodic grouping and sentence type effects, the final pitch contour can be determined. High intelligibility and acceptable naturalness of the synthesized speech have been confirmed by subjective listening tests.


Full Paper

Bibliographic reference.  Abutalebi, Hamid Reza / Bijankhan, Mahmood (2000): "Implementation of a text-to-speech system for farsi language", In ICSLP-2000, vol.1, 661-664.