Neural Network Program Package for Prosody Modeling

Neural Network Program Package for Prosody Modeling

This contribution describes the programme for one part of theautomatic Text-to-Speech (TTS) synthesis. Some experiments (for example[14]) documented the considerable improvement of the naturalness ofsynthetic speech, but this approach requires completing the inputfeature values by hand. This complet...

Full description

Bibliographic Details
Main Authors: J. Santarius, J. Tuckova
Format: Article
Language:English
Published: Spolecnost pro radioelektronicke inzenyrstvi 2004-04-01
Series:Radioengineering
Online Access:http://www.radioeng.cz/fulltexts/2004/04_01_17_21.pdf
id doaj-07bd22af81864e9d8561b704788e7926
record_format Article
spelling doaj-07bd22af81864e9d8561b704788e79262020-11-24T22:38:43ZengSpolecnost pro radioelektronicke inzenyrstviRadioengineering1210-25122004-04-011311721Neural Network Program Package for Prosody ModelingJ. SantariusJ. TuckovaThis contribution describes the programme for one part of theautomatic Text-to-Speech (TTS) synthesis. Some experiments (for example[14]) documented the considerable improvement of the naturalness ofsynthetic speech, but this approach requires completing the inputfeature values by hand. This completing takes a lot of time for bigfiles. We need to improve the prosody by other approaches which useonly automatically classified features (input parameters). Theartificial neural network (ANN) approach is used for the modeling ofprosody parameters. The program package contains all modules necessaryfor the text and speech signal pre-processing, neural network training,sensitivity analysis, result processing and a module for the creationof the input data protocol for Czech speech synthesizer ARTIC [1].www.radioeng.cz/fulltexts/2004/04_01_17_21.pdf
collection DOAJ
language English
format Article
sources DOAJ
author J. Santarius
J. Tuckova
spellingShingle J. Santarius
J. Tuckova
Neural Network Program Package for Prosody Modeling
Radioengineering
author_facet J. Santarius
J. Tuckova
author_sort J. Santarius
title Neural Network Program Package for Prosody Modeling
title_short Neural Network Program Package for Prosody Modeling
title_full Neural Network Program Package for Prosody Modeling
title_fullStr Neural Network Program Package for Prosody Modeling
title_full_unstemmed Neural Network Program Package for Prosody Modeling
title_sort neural network program package for prosody modeling
publisher Spolecnost pro radioelektronicke inzenyrstvi
series Radioengineering
issn 1210-2512
publishDate 2004-04-01
description This contribution describes the programme for one part of theautomatic Text-to-Speech (TTS) synthesis. Some experiments (for example[14]) documented the considerable improvement of the naturalness ofsynthetic speech, but this approach requires completing the inputfeature values by hand. This completing takes a lot of time for bigfiles. We need to improve the prosody by other approaches which useonly automatically classified features (input parameters). Theartificial neural network (ANN) approach is used for the modeling ofprosody parameters. The program package contains all modules necessaryfor the text and speech signal pre-processing, neural network training,sensitivity analysis, result processing and a module for the creationof the input data protocol for Czech speech synthesizer ARTIC [1].
url http://www.radioeng.cz/fulltexts/2004/04_01_17_21.pdf
work_keys_str_mv AT jsantarius neuralnetworkprogrampackageforprosodymodeling
AT jtuckova neuralnetworkprogrampackageforprosodymodeling
_version_ 1725712273212702720

Similar Items