Modular Architecture for Research on speech sYnthesis

Modular Architecture for Research on speech sYnthesis

  • Duration:

MARY is a Text-to-Speech Synthesis System for German, English and Tibetan. It was originally developed as a collaborative project of DFKI's language technology lab and the Institute of Phonetics at Saarland University and is now being maintained by DFKI.

MARY is designed to be highly modular, with a special focus on transparency and accessibility of intermediate processing steps. This makes it a suitable tool for Research and Development.

Publications about the project

Sébastien Le Maguer; Ingmar Steiner; Francesco Tombini; Pradipta Deb; Moitree Basu; Insa Kröger

In: Blizzard Challenge. SynSIG Blizzard Challenge (SSW-2018), located at Interspeech, September 8, Hyderabad, India, ISCA, 9/2018.

To the publication

Eran Raveh; Ingmar Steiner; Iona Gessinger; Bernd Möbius

In: Alexey Karpov; Oliver Jokisch; Rodmonga Potapova (Hrsg.). 20th International Conference on Speech and Computer (SPECOM). Pages 552-562, Lecture Notes in Artificial Intelligence (LNAI), Vol. 11096, Springer, 9/2018.

To the publication

Iona Gessinger; Antje Schweitzer; Bistra Andreeva; Eran Raveh; Bernd Möbius; Ingmar Steiner

In: Katarzyna Klessa; Jolanta Bachan; Agnieszka Wagner; Maciej Karpiński; Daniel Śledziński (Hrsg.). 9th International Conference on Speech Prosody. International Conference on Speech Prosody (Speech Prosody-2018), June 13-16, Poznan, Poland, Pages 225-229, ISCA, 6/2018.

To the publication

German Research Center for Artificial Intelligence
Deutsches Forschungszentrum für Künstliche Intelligenz