System for text to speech synthesis

Lambe Čočorovski (2013) System for text to speech synthesis. EngD thesis.

Preview

Abstract

Speech synthesis is increasingly developing as the computers are getting more powerful. In this work, the area of speech synthesis is introduced through a diphone and unit selection based synthesis methods. The basic architecture of the synthesizers, text analysis and the use of algorithms for digital signal processing are described. DTW algorithm was used as a tool for labelling phonemes from an audio signal using MFCC coefficients and LPC coding of the speech signal. Festival Speech Synthesis System with Edinburgh Speech Tools and Festvox documentation are used for building new languages. The characteristics of the Slovenian language are described and their usage in the Festival. We have successfully built two speech synthesisers and the results are evaluated using a MOS scale.

Item Type:

Thesis (EngD thesis)

Keywords:

speech synthesis, Festival, Festvox, diphone synthesis, unit selection

Number of Pages:

Language of Content:

English

Mentor / Comentors:

Name and Surname	ID	Function
prof. dr. Dušan Kodek	236	Mentor

Link to COBISS:

http://www.cobiss.si/scripts/cobiss?command=search&base=50070&select=(ID=9962580)

Institution:

University of Ljubljana

Department:

Faculty of Computer and Information Science

Item ID:

2073

Date Deposited:

28 Jun 2013 16:55

Last Modified:

16 Jul 2013 14:03

URI:

http://eprints.fri.uni-lj.si/id/eprint/2073

Actions (login required)

View Item