Evaluation of a Multilingual Tts System with Respect to the Prosodic Quality

  title={Evaluation of a Multilingual Tts System with Respect to the Prosodic Quality},
  author={R{\"u}diger Hoffmann and Diane Hirschfeld and Oliver Jokisch and Ulrich Kordon and Hansj{\"o}rg Mixdorff and Dieter Mehnert},
Improving the naturalness of synthetic speech is an essential task in developing a text-to-speech (TTS) system. Mainly, it depends on the quality of the prosody model which is utilized in the TTS system. For our TTS system called DreSS (Dresden Speech Synthesizer), we compared three different met hods for generating the F0 contour to each other as well as to other synthesizers. Natural speech samples were used as a reference. Results show, that on a naturalness scale from 0 to 4, the natural… CONTINUE READING


Publications citing this paper.


Publications referenced by this paper.
Showing 1-10 of 11 references

Comparing the naturalness of several approaches for generating f0 contours in German text-to-speech systems

  • H. Mixdorff, D. Mehnert, D. Hirschfeld
  • Proc ASA
  • 1999
1 Excerpt

Multilingual speech synthesis for car applications

  • R. Hoffmann, U. Kordon, H. J. Holland, S. Netz
  • Proc. ISATA ’99,
  • 1999
1 Excerpt

Intonation Patterns of German – Quantitative Analysis and Synthesis of F0 Contours

  • H. Mixdorff
  • 1998
1 Excerpt

Perceptual evaluation of three different approaches for generating F0 contours in TTS

  • H. Mixdorff, D. Mehnert
  • Fortschritte der Akustik 1998, Zürich,
  • 1998
1 Excerpt

Modeling the process of fundamental frequency control of speech for synthesis of tonal features of various languages

  • H. Fujisaki
  • China-Japan Symposium on Advanced Information…
  • 1997
1 Excerpt

Similar Papers

Loading similar papers…