A complete text-to-speech synthesis system in Tamil

Author

Rama, G. L Jayavardhana ; Ramakrishnan, A.G. ; Muralishankar, R. ; Prathibha, R.

Author_Institution

Dept. of Electr. Eng., Indian Inst. of Sci., Bangalore, India

fYear

2002

fDate

11-13 Sept. 2002

Firstpage

191

Lastpage

194

Abstract

We report the design and development of Thirukkural, the first text-to-speech converter in Tamil. Syllables of different lengths have been selected as units since Tamil is a syllabic language. An automatic segmentation algorithm has been devised for segmenting syllables into consonant and vowel. The units are pitch marked using the discrete cosine transform-spectral autocorrelation function (DCTSAF). Prosodic information is captured in tables based on extensive observation of spoken Tamil. During synthesis, DCT based pitch modification is applied for both waveform interpolation and modifying pitch contour for different sentence modalities. Thirukkural is designed in VC++ and runs on Windows 95/98/NT. Perceptual evaluation by natives show that the synthesized speech is intelligible and fairly natural.

Keywords

correlation methods; discrete cosine transforms; interpolation; spectral analysis; speech intelligibility; speech processing; speech synthesis; DCTSAF; Tamil syllabic language; Thirukkural; VC++; automatic segmentation algorithm; discrete cosine transform-spectral autocorrelation function; perceptual evaluation; pitch contour modification; pitch modification; prosodic information; speech intelligibility; speech naturalness; text-to-speech synthesis system; unit pitch marking; waveform interpolation; Autocorrelation; Computer interfaces; Discrete cosine transforms; Discrete transforms; Flowcharts; Interpolation; Natural languages; Speech analysis; Speech synthesis; User interfaces;

fLanguage

English

Publisher

ieee

Conference_Titel

Speech Synthesis, 2002. Proceedings of 2002 IEEE Workshop on

Print_ISBN

0-7803-7395-2

Type

conf

DOI

10.1109/WSS.2002.1224406

Filename

1224406