Neuro-fuzzy Filtering Techniques for the Analysis of Complex Acoustic Scenarios
A neuro-fuzzy based filter for acoustic filtering is described. In order to perform temporal filtering, a proper architecture was developed exploiting a buffer memory. Subband analysis was used. This led to an architecture composed of a QMF filter bank, a neuro-fuzzy network for each subband and a reconstruction QMF filter bank Many simulation results are described relative to signals captured from real acoustic scenarios.
KeywordsFilter Bank Flicker Noise Pink Noise Random Search Algorithm Auditory Scene Analysis
Unable to display preview. Download preview PDF.
- D. P. W. Ellis, Prediction driven computational auditory scene analysis, submitted to MIT in fulfillment of the requirements for the degree of Doctor of Philosophy in Electrical Engineering, June 1996.Google Scholar
- K. Arakawa, “Fuzzy Rule Based Signal Processing and its applications to Image Restoration” IEEE Journal on Selected Areas in Communications, Vol. 12, no.9, December 94.Google Scholar
- N. Hess-Nielsen, M. V. Wickerhauser, “Wavelets and time frequency analysis”, Proc. IEEE, Vol. 84, No. 4, April 1996.Google Scholar
- J.M.Morris and R. Perevali, “Minimum-bandwidth discrete-time wavelets”, Signal Processing, Elsevier, vol 76, 1999.Google Scholar
- P.L.Ainsleigh and C. Chui,“A B-Wavelet-Based Noise Reduction Algorithm”, IEEE Transactions on Signal Processing, Vol. 4, No.5, May 96.Google Scholar
- D. Nauf, F. Klawonn, R. Kruse, Foundations of Neuro-Fuzzy Systems, Wiley, 1996.Google Scholar