Noncausal all-pole modeling of voiced speech

Author

Gardner, William R. ; Rao, Bhaskar D.

Author_Institution

Qualcomm Inc., San Diego, CA, USA

Volume

5

Issue

1

fYear

1997

fDate

1/1/1997 12:00:00 AM

Firstpage

1

Lastpage

10

Abstract

This paper introduces noncausal all-pole models that are capable of efficiently capturing both the magnitude and phase information of voiced speech. It is shown that noncausal all-pole filter models are better able to match both magnitude and phase information and are particularly appropriate for voiced speech due to the nature of the glottal excitation. By modeling speech in the frequency domain, the standard difficulties that occur when using noncausal all-pole filters are avoided. Several algorithms for determining the model parameters based on frequency-domain information and the masking effects of the ear are described. Our work suggests that high-quality voiced speech can be produced using a 14th-order noncausal all-pole model

Keywords

amplitude estimation; filtering theory; frequency-domain analysis; phase estimation; poles and zeros; speech intelligibility; speech processing; ear; frequency-domain information; glottal excitation; high quality voiced speech; magnitude information; masking effects; model parameters; noncausal all-pole filter models; noncausal all-pole modeling; parameter estimation algorithms; phase information; speech synthesis; voiced speech; Ear; Frequency domain analysis; Information filtering; Information filters; Linear predictive coding; Matched filters; Predictive models; Speech coding; Transfer functions; Vocoders;

fLanguage

English

Journal_Title

Speech and Audio Processing, IEEE Transactions on

Publisher

ieee

ISSN

1063-6676

Type

jour

DOI

10.1109/89.554263

Filename

554263