Open data for speech synthesis of Austrian German language varieties

by Michael Pucher, Michaela Rausch-Supola, Sylvia Moosmüller, Markus Toman, Dietmar Schabus, Friedrich Neubarth
Abstract:
In this paper we summarize open data sets and open source software that we have released for Austrian German language varieties as a result of several research projects. We also describe some data sets that are released for research purposes only, due to licensing limitations. From the development of these resources we draw conclusions concerning the collection and licensing of such data with a special focus on the problem of speech synthesis where the voice identity of the speaker plays an important role. Furthermore we discuss recordings that we plan to perform in the future, where we aim to cover most Austrian dialects.
Reference:
Michael Pucher, Michaela Rausch-Supola, Sylvia Moosmüller, Markus Toman, Dietmar Schabus, Friedrich Neubarth, “Open data for speech synthesis of Austrian German language varieties”, In 12. Tagung Phonetik und Phonologie im deutschsprachigen Raum, Munich, Germany, 2016.
Bibtex Entry:
@InProceedings{Pucher2016,
  Title                    = {Open data for speech synthesis of Austrian German language varieties},
  Author                   = {Michael Pucher and Michaela Rausch-Supola and Sylvia Moosmüller and Markus Toman and Dietmar Schabus and Friedrich Neubarth},
  Booktitle                = {12. Tagung Phonetik und Phonologie im deutschsprachigen Raum},
  Year                     = {2016},

  Address                  = {Munich, Germany},
  Month                    = oct,

  Abstract                 = {In this paper we summarize open data sets and open source software that we have released for Austrian German language varieties as a result of several research projects. We also describe some data sets that are released for research purposes only, due to licensing limitations. From the development of these resources we draw conclusions concerning the collection and licensing of such data with a special focus on the problem of speech synthesis where the voice identity of the speaker plays an important role. Furthermore we discuss recordings that we plan to perform in the future, where we aim to cover most Austrian dialects.},
}