Title :
A Syllable-Phoneme Solution for Efficient Farsi Text-to-Speech Systems
Author :
Rasekh, Ehsan ; Eshghi, Mohammad
Author_Institution :
Electr. & Comput. Eng. Fac., Shahid Beheshti Univ., Tehran
Abstract :
In this paper a new syllable-phoneme database approach for text to phoneme conversion in Farsi text to speech systems is presented. The volume of the database in this method is about 15% of a convenient system using a word-phoneme lexicon. This method shows better results comparing to word-phoneme lexicon and rule based text-to-phoneme systems in Farsi. With training the system over 50% to 100% of a 2024 word database and testing over the entire database, the performance of 69.43% to 96.06% is achieved
Keywords :
natural languages; speech processing; speech synthesis; Farsi; syllable-phoneme database approach; text to phoneme conversion; text-to-speech system; word-phoneme lexicon; Artificial neural networks; Costs; Data engineering; Databases; Natural languages; Optical computing; Shape; Speech synthesis; System testing; Vocabulary;
Conference_Titel :
TENCON 2006. 2006 IEEE Region 10 Conference
Conference_Location :
Hong Kong
Print_ISBN :
1-4244-0548-3
Electronic_ISBN :
1-4244-0549-1
DOI :
10.1109/TENCON.2006.343933
