Automatic statistical evaluation of quality of unit selection speech synthesis with different prosody manipulations

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F20%3A43958604" target="_blank" >RIV/49777513:23520/20:43958604 - isvavai.cz</a>

Výsledek na webu

<a href="http://iris.elf.stuba.sk/JEEEC/data/pdf/2_120-02.pdf" target="_blank" >http://iris.elf.stuba.sk/JEEEC/data/pdf/2_120-02.pdf</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.2478/jee-2020-0012" target="_blank" >10.2478/jee-2020-0012</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Automatic statistical evaluation of quality of unit selection speech synthesis with different prosody manipulations

Popis výsledku v původním jazyce

Quality of speech synthesis is a crucial issue in comparison of various text-to-speech (TTS) systems. We proposed a system for automatic evaluation of speech quality by statistical analysis of temporal features (time duration, phrasing, and time structuring of an analysed sentence) together with standard spectral and prosodic features. This system was successfully tested on sentences produced by a unit selection speech synthesizer with a male as well as a female voice using two different approaches to prosody manipulation. Experiments have shown that for correct, sharp, and stable results all three types of speech features (spectral, prosodic, and temporal) are necessary. Furthermore, the number of used statistical parameters has a significant impact on the correctness and precision of the evaluated results. It was also demonstrated that the stability of the whole evaluation process is improved by enlarging the used speech material. Finally, the functionality of the proposed system was verified by comparison of the results with those of the standard listening test.

Název v anglickém jazyce

Automatic statistical evaluation of quality of unit selection speech synthesis with different prosody manipulations

Popis výsledku anglicky

Quality of speech synthesis is a crucial issue in comparison of various text-to-speech (TTS) systems. We proposed a system for automatic evaluation of speech quality by statistical analysis of temporal features (time duration, phrasing, and time structuring of an analysed sentence) together with standard spectral and prosodic features. This system was successfully tested on sentences produced by a unit selection speech synthesizer with a male as well as a female voice using two different approaches to prosody manipulation. Experiments have shown that for correct, sharp, and stable results all three types of speech features (spectral, prosodic, and temporal) are necessary. Furthermore, the number of used statistical parameters has a significant impact on the correctness and precision of the evaluated results. It was also demonstrated that the stability of the whole evaluation process is improved by enlarging the used speech material. Finally, the functionality of the proposed system was verified by comparison of the results with those of the standard listening test.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

20205 - Automation and control systems

Projekt

<a href="/cs/project/GA19-19324S" target="_blank" >GA19-19324S: Plně trénovatelná syntéza české řeči z textu s využitím hlubokých neuronových sítí</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2020

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Journal of ELECTRICAL ENGINEERING

ISSN

1335-3632

e-ISSN

—

Svazek periodika

71

Číslo periodika v rámci svazku

2

Stát vydavatele periodika

SK - Slovenská republika

Počet stran výsledku

9

Strana od-do

78-86

Kód UT WoS článku

000536287900002

EID výsledku v databázi Scopus

2-s2.0-85085749611