• DocumentCode
    3239980
  • Title

    Neural speech enhancement in the time-frequency domain

  • Author

    Volkmer, Markus

  • Author_Institution
    Dept. of Comput. Eng. VI, Technische Univ. Hamburg-Harburg, Hamburg, Germany
  • fYear
    2003
  • fDate
    17-19 Sept. 2003
  • Firstpage
    617
  • Lastpage
    626
  • Abstract
    A pulsed neural network approach for the enhancement of speech is presented. It employs a recently introduced pulse-induced masking principle on a joint time-frequency representation of signals. Based on the notion of a spectro-temporal receptive field, the principle provides signal-dependent dictionary elements, which are integrated into an enhancement process in the time-frequency domain. Experiments on signals with different types of noise are conducted with the resulting neural time-frequency domain method.
  • Keywords
    neural nets; speech enhancement; time-frequency analysis; neural speech enhancement; neural time-frequency domain method; pulse-induced masking principle; pulsed neural network approach; signal-dependent dictionary elements; spectrotemporal receptive field; Artificial neural networks; Dictionaries; Intelligent networks; Neural networks; Signal processing; Signal to noise ratio; Speech analysis; Speech enhancement; Speech processing; Time frequency analysis;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Neural Networks for Signal Processing, 2003. NNSP'03. 2003 IEEE 13th Workshop on
  • ISSN
    1089-3555
  • Print_ISBN
    0-7803-8177-7
  • Type

    conf

  • DOI
    10.1109/NNSP.2003.1318061
  • Filename
    1318061