Title :
Speech signal reconstruction based on higher order spectra
Author :
Seetharaman, S. ; Jernigan, M.E.
Author_Institution :
Dept. of Syst. Design Eng., Waterloo Univ., Ont., Canada
Abstract :
The need for the use of higher order spectra, namely, bispectra in speech signal analysis is discussed. It is shown that speech signal can be reconstructed from its bispectral information. The results suggest that bispectrum can be a useful analysis tool for gaining insight into the speech production process
Keywords :
speech analysis and processing; bispectra; higher order spectra; speech production process; speech reconstruction; speech signal analysis; Filtering theory; Fourier transforms; Higher order statistics; Nonlinear filters; Predictive models; Signal processing; Signal reconstruction; Speech processing; Speech synthesis; Statistical distributions;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1988. ICASSP-88., 1988 International Conference on
Conference_Location :
New York, NY
DOI :
10.1109/ICASSP.1988.196683
