VITS: Quality vs. Speed Analysis

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F23%3A43969619" target="_blank" >RIV/49777513:23520/23:43969619 - isvavai.cz</a>

Výsledek na webu

<a href="https://link.springer.com/chapter/10.1007/978-3-031-40498-6_19" target="_blank" >https://link.springer.com/chapter/10.1007/978-3-031-40498-6_19</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-031-40498-6_19" target="_blank" >10.1007/978-3-031-40498-6_19</a>

Jazyk výsledku

angličtina

Název v původním jazyce

VITS: Quality vs. Speed Analysis

Popis výsledku v původním jazyce

In this paper, we analyze the performance of a modern end-to-end speech synthesis model called Variational Inference with adversarial learning for end-to-end Text-to-Speech (VITS). We build on the original VITS model and examine how different modifications to its architecture affect synthetic speech quality and computational complexity. Experiments with two Czech voices, a male and a female, were carried out. To assess the quality of speech synthesized by the different modified models, MUSHRA listening tests were performed. The computational complexity was measured in terms of synthesis speed over real time. While the original VITS model is still preferred regarding speech quality, we present a modification of the original structure with a significantly better response yet providing acceptable output quality. Such a configuration can be used when system response latency is critical.

Název v anglickém jazyce

VITS: Quality vs. Speed Analysis

Popis výsledku anglicky

In this paper, we analyze the performance of a modern end-to-end speech synthesis model called Variational Inference with adversarial learning for end-to-end Text-to-Speech (VITS). We build on the original VITS model and examine how different modifications to its architecture affect synthetic speech quality and computational complexity. Experiments with two Czech voices, a male and a female, were carried out. To assess the quality of speech synthesized by the different modified models, MUSHRA listening tests were performed. The computational complexity was measured in terms of synthesis speed over real time. While the original VITS model is still preferred regarding speech quality, we present a modification of the original structure with a significantly better response yet providing acceptable output quality. Such a configuration can be used when system response latency is critical.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

20205 - Automation and control systems

Projekt

<a href="/cs/project/TL05000546" target="_blank" >TL05000546: Využití multimediálního výkladového slovníku pro moderní výuku češtiny</a><br>

Návaznosti

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

Rok uplatnění

2023

Kód důvěrnosti údajů

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

Název statě ve sborníku

Text, Speech, and Dialogue 26th International Conference, TSD 2023, Pilsen, Czech Republic, September 4–6, 2023, Proceedings

ISBN

978-3-031-40497-9

ISSN

0302-9743

e-ISSN

1611-3349

Počet stran výsledku

12

Strana od-do

214-225

Název nakladatele

Springer International Publishing

Místo vydání

Cham

Místo konání akce

Pilsen, Czech Republic

Datum konání akce

4. 9. 2023

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—