Advertisement

ICANN ’93 pp 422-422 | Cite as

Modulation-Frequency Encoding of Speech with Applications to Neural Speech Recognizers

  • M. Paping
  • H. W. Strube
  • T. Gramss
Conference paper

Abstract

Most of the present speech recognition systems make use of a signal representation based on parameter frames equidistant in time. We propose a new method of signal representation which is motivated by linguistic, psychoacoustic and neurobiological facts.

Keywords

Feature Vector Signal Representation Continuous Speech Speech Segment Female Speaker 
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.

Copyright information

© Springer-Verlag London Limited 1993

Authors and Affiliations

  • M. Paping
    • 1
  • H. W. Strube
    • 1
  • T. Gramss
    • 1
  1. 1.Drittes Physikalisches InstitutUniversity of GöttingenGöttingenGermany

Personalised recommendations