Share Email Print
cover

Proceedings Paper

Physical properties of modification of speech signal fragments
Author(s): Mikhail N. Gusev
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

The methods used for modification of separate speech signals fragments in the process of speech synthesis by arbitrary text are described in this report. Three groups of sounds differ in the modification methods of frequency characteristics. Two groups of sounds differ in that they need different methods of duration changes. To modify the samples of a speaker's voice by the methods used it is necessary to make pre-marking, so called segementation. As variable speech fragments, the allophones are taken. The modification methods described allow form arbitrary speech successions in the wide intonation diapason on the basis of limited amount of the speaker's voice patterns.

Paper Details

Date Published: 5 April 2004
PDF: 6 pages
Proc. SPIE 5400, Seventh International Workshop on Nondestructive Testing and Computer Simulations in Science and Engineering, (5 April 2004); doi: 10.1117/12.555540
Show Author Affiliations
Mikhail N. Gusev, St. Petersburg State Univ. of Telecommunications (Russia)


Published in SPIE Proceedings Vol. 5400:
Seventh International Workshop on Nondestructive Testing and Computer Simulations in Science and Engineering
Alexander I. Melker, Editor(s)

© SPIE. Terms of Use
Back to Top