Text-To-Speech Engine with Austrian German Corpus

Christian Kranzler; Franz Pernkopf; Rudolf Muhr; Michael Pucher; Friedrich Neubarth

Text-To-Speech Engine with Austrian German Corpus

Christian Kranzler
, Franz Pernkopf
, Rudolf Muhr
, Michael Pucher
, Friedrich Neubarth

Institute of Signal Processing and Speech Communication (4420)

Research output: Chapter in Book/Report/Conference proceeding › Conference paper › peer-review

Abstract

This paper deals with developing a unit selection speech corpus for the Austrian variety of German by (re)using the resources for German and adapt them to Austrian German. This means adaptation on different levels such as lexicon level, phone level, or speech data level, whereas a compromise between reusing the given resource and an exact time-consuming phonetic transcription has to be found. In experiments our Austrian German voice was evaluated by speech experts against the correctness of pronunciation while using an adapted Austrian lexicon.

Original language	English
Title of host publication	International Conference on Speech and Computer, SPECOM
Number of pages	4
Publication status	Published - 2009
Event	International Conference on Speech and Computer: SPECOM 2009 - St. Petersburg, Russian Federation Duration: 21 Jun 2009 → 25 Jun 2009

Conference

Conference	International Conference on Speech and Computer
Country/Territory	Russian Federation
City	St. Petersburg
Period	21/06/09 → 25/06/09

Access to Document

SPECOM2009_KranzlerPernkopfMuhrPucherNeubarth.pdfFinal published version, 77.2 KB

Cite this

@inproceedings{7f77777e4c96485c9fe028cd9aa5d367,

title = "Text-To-Speech Engine with Austrian German Corpus",

abstract = "This paper deals with developing a unit selection speech corpus for the Austrian variety of German by (re)using the resources for German and adapt them to Austrian German. This means adaptation on different levels such as lexicon level, phone level, or speech data level, whereas a compromise between reusing the given resource and an exact time-consuming phonetic transcription has to be found. In experiments our Austrian German voice was evaluated by speech experts against the correctness of pronunciation while using an adapted Austrian lexicon.",

author = "Christian Kranzler and Franz Pernkopf and Rudolf Muhr and Michael Pucher and Friedrich Neubarth",

year = "2009",

language = "English",

booktitle = "International Conference on Speech and Computer, SPECOM",

note = "International Conference on Speech and Computer : SPECOM 2009 ; Conference date: 21-06-2009 Through 25-06-2009",

}

TY - GEN

T1 - Text-To-Speech Engine with Austrian German Corpus

AU - Kranzler, Christian

AU - Pernkopf, Franz

AU - Muhr, Rudolf

AU - Pucher, Michael

AU - Neubarth, Friedrich

PY - 2009

Y1 - 2009

N2 - This paper deals with developing a unit selection speech corpus for the Austrian variety of German by (re)using the resources for German and adapt them to Austrian German. This means adaptation on different levels such as lexicon level, phone level, or speech data level, whereas a compromise between reusing the given resource and an exact time-consuming phonetic transcription has to be found. In experiments our Austrian German voice was evaluated by speech experts against the correctness of pronunciation while using an adapted Austrian lexicon.

AB - This paper deals with developing a unit selection speech corpus for the Austrian variety of German by (re)using the resources for German and adapt them to Austrian German. This means adaptation on different levels such as lexicon level, phone level, or speech data level, whereas a compromise between reusing the given resource and an exact time-consuming phonetic transcription has to be found. In experiments our Austrian German voice was evaluated by speech experts against the correctness of pronunciation while using an adapted Austrian lexicon.

M3 - Conference paper

BT - International Conference on Speech and Computer, SPECOM

T2 - International Conference on Speech and Computer

Y2 - 21 June 2009 through 25 June 2009

ER -

Text-To-Speech Engine with Austrian German Corpus

Abstract

Conference

Access to Document

Fingerprint

Cite this