Examining the Ability of One–Class Classifier to Ensure the Spectral Smoothness of Concatenated Units

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F16%3A43929966" target="_blank" >RIV/49777513:23520/16:43929966 - isvavai.cz</a>

Výsledek na webu

<a href="http://dx.doi.org/10.1109/ICSP.2016.7877899" target="_blank" >http://dx.doi.org/10.1109/ICSP.2016.7877899</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/ICSP.2016.7877899" target="_blank" >10.1109/ICSP.2016.7877899</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Examining the Ability of One–Class Classifier to Ensure the Spectral Smoothness of Concatenated Units

Popis výsledku v původním jazyce

We present initial experiments with one–class clas- sification, aimed at replacing the “classic” heuristics–based mea- sures used to estimate the smoothness of units concatenated together within unit selection speech synthesizers. A set of spectral feature distances was computed between neighbouring frames in natural speech recordings, i.e. those representing natural joins, from which the per–vowel classifier was trained. For the evaluation, we carried out ad–hoc listening tests collecting several examples of smooth and discontinuous joins, against which the classifier is tested. In addition, we also plugged the classifier into our TTS system to verify that the technique is capable of replacing the classic approach in the generic unit selection procedure.

Název v anglickém jazyce

Examining the Ability of One–Class Classifier to Ensure the Spectral Smoothness of Concatenated Units

Popis výsledku anglicky

We present initial experiments with one–class clas- sification, aimed at replacing the “classic” heuristics–based mea- sures used to estimate the smoothness of units concatenated together within unit selection speech synthesizers. A set of spectral feature distances was computed between neighbouring frames in natural speech recordings, i.e. those representing natural joins, from which the per–vowel classifier was trained. For the evaluation, we carried out ad–hoc listening tests collecting several examples of smooth and discontinuous joins, against which the classifier is tested. In addition, we also plugged the classifier into our TTS system to verify that the technique is capable of replacing the classic approach in the generic unit selection procedure.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

20205 - Automation and control systems

Projekt

<a href="/cs/project/LO1506" target="_blank" >LO1506: Podpora udržitelnosti centra NTIS - Nové technologie pro informační společnost</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2016

Kód důvěrnosti údajů

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

Název statě ve sborníku

2016 IEEE 13th International Conference on Signal Processing Proceedings

ISBN

978-1-5090-1345-6

ISSN

—

e-ISSN

—

Počet stran výsledku

6

Strana od-do

578-583

Název nakladatele

IEEE

Místo vydání

New York

Místo konání akce

Chengdu, China

Datum konání akce

6. 11. 2016

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000406056300115