| Autor: |
Nguyen Thi Bich Diep, Nguyen Van Tao, Pham Thi Mai Huong, Phung Thi Thu Hien, Phung Trung Nghia |
| Rok vydání: |
2016 |
| Předmět: |
|
| Zdroj: |
REV Journal on Electronics and Communications. |
| ISSN: |
1859-378X |
| DOI: |
10.21553/rev-jec.106 |
| Popis: |
The articulators typically move smoothly during speech production. Therefore, speech features of natural speech are generally smooth. However, over-smooth causes the “muffleness" and the reduction in identification emotions / expressions / styles in synthesized speech that can affect to the perception of the naturalness in synthesized speech. In the literature, statistical variances of static spectral features have been used as a measure of smoothness in synthesized speech but they are not sufficient enough. This paper aims to propose a speech smoothness measure that can be efficiently applied to evaluate the smoothness of synthesized speech. Experiments show that the proposed measures are reliable and efficient to measure smoothness of different kinds of synthesized speech. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
