Prosodic Features in German Speech: Stress Assignment by Man and Machine

  • E. Nöth
  • H. Niemann
  • S. Schmölz
Conference paper
Part of the NATO ASI Series book series (volume 46)


We present a method for automatically assigning a stress score to every syllable in an utterance. The syllables are detected by looking at the energy in three energy bands. Based on features representing the prosodic parameters duration, intensity, and pitch, a stress assignment number for each syllable is calculated. The algorithm is tested on material collected from real dialogs. The automatic stress assignment is compared with the result of a stress perception experiment by human listeners.


Stress Score Read Sentence Stress Assignment Unstressed Syllable Prosodic Parameter 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [DIK 81a]
    Dik, S. G: “Functional Grammar”, Foris Publications, Dordrecht, 1981.Google Scholar
  2. [EHR 88a]
    Ehrlich, U., Niemann, H.: “Using Semantic and Pragmatic Knowledge for the Interpretation of Syntactic Constituents”, in this volume.Google Scholar
  3. Kunzmann, S., Kuhn, T., Niemann, H.: “An Experimental Environment for Generating Word Hypotheses in Continuous Speech”, in this volume.Google Scholar
  4. [LEA 80a]
    Lea, W.: “Prosodic Aids to Speech Recognition”, in Lea, W.: “Trends in Speech Recognition”, Prentice Hall, N. J., 1980.Google Scholar
  5. [NIE 84a]
    Niemann, H., Brietzmann, A., Mühlfeld, R., Regel, P., Schukat, G.: “The Speech Understanding and Dialog System EVAR”, in De Mori, R., Suen, G: “New Systems and Architectures for Automatic Speech Recognition and Synthesis”, Springer Verlag, Berlin, 1985.Google Scholar
  6. [SCH 87a]
    Schmölz, S.: “Zur automatischen Bestimmung der Satzbetonung”, Diplomarbeit, Lehrstuhl für Informatik 5 (Mustererkennung), Universität Erlangen, 1987.Google Scholar
  7. [SEN 78a]
    [SENSennef, S.: “Real-time harmonic pitch detector”, IEEE Trans. Acoust, Speech, Signal Processing, vol. ASSP-26, S. 358–364, 1978.CrossRefGoogle Scholar
  8. [VAI82a]
    Vaissiere, J.: “A Suprasegmental Component in a French Speech Recognition System: Reducing the Number of Lexical Hypotheses and Detecting the Main Boundary”, Recherches/Acoustique, CNET, Vol. 7, 1982/83.Google Scholar
  9. [WAI 86a]
    Waibel, A.: “Prosody and Speech Recognition”, Ph.D. thesis, Carnegie-Mellon University, Pittsburgh, USA, 1986.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1988

Authors and Affiliations

  • E. Nöth
    • 1
  • H. Niemann
    • 1
  • S. Schmölz
    • 1
  1. 1.Lehrstuhl für Informatik 5 (Mustererkennung)Friedrich-Alexander-Universität Erlangen-NürnbergErlangenF.R. Germany

Personalised recommendations