Diphone preparation for Bangla text to speech synthesis

Author

Rashid, Mohammad M. ; Hussain, Muhammad Awais ; Rahman, Md Saifur

Author_Institution

Shahjalal Univeristy of Sci. & Technol., Sylhet, Bangladesh

fYear

2009

fDate

21-23 Dec. 2009

Firstpage

226

Lastpage

230

Abstract

This paper presents methodologies involved in diphone preparation for Bangla text to speech synthesis. A concatenation based synthesis system comprises basically two modules- one is natural language processing and other is digital signal processing (DSP). Natural language processing implies converting text to its pronounceable text, called text normalization and the diphone selection method based on the normalized text is called Graphene to Phoneme (G2P) conversion. We developed a speech synthesizer for Bangla using diphone based concatenative approach. Diphone preparation, labeling and selection techniques are described in this paper.

Keywords

natural language processing; speech synthesis; text analysis; Bangla text; concatenation based synthesis system; digital signal processing; diphone preparation; diphone selection method; graphene conversion; natural language processing; normalized text; phoneme conversion; pronounceable text; speech synthesis; speech synthesizer; text normalization; Cleaning; Computer science; Data engineering; Data mining; Databases; Detection algorithms; Information technology; Sorting; Speech synthesis; diphone; grapheme-to-phoneme; speech synthesis; text normalization;

fLanguage

English

Publisher

ieee

Conference_Titel

Computers and Information Technology, 2009. ICCIT '09. 12th International Conference on

Conference_Location

Dhaka

Print_ISBN

978-1-4244-6281-0

Type

conf

DOI

10.1109/ICCIT.2009.5407135

Filename

5407135