Particle Filter Inference in an Articulatory-Based Speech Model

Author

Beierholm, Thomas ; Winther, Ole

Author_Institution

GN ReSound A/S, Ballerup

Volume

14

Issue

11

fYear

2007

Firstpage

883

Lastpage

886

Abstract

A speech model parameterized by formant frequencies, formant bandwidths, and formant gains is proposed. Inference in the model is made by particle filtering for the application of speech enhancement. The advantage of the proposed parameterization over existing parameterizations based on auto-regressive (AR) coefficients or reflection coefficients is the smooth time-varying behavior of the parameters and their loose coupling. Experiments confirm this advantage both in terms of parameter estimation and SNR improvement.

Keywords

autoregressive processes; particle filtering (numerical methods); speech enhancement; articulatory-based speech model; auto-regressive coefficient; formant bandwidth; formant frequency; formant gain; loose coupling; parameter estimation; particle filter inference; smooth time-varying parameter behavior; speech enhancement; Bandwidth; Filtering; Frequency; Particle filters; Reflection; Signal processing; Speech enhancement; Speech processing; Stability; State estimation; Formant frequency; particle filtering; time-varying auto-regressive speech model;

fLanguage

English

Journal_Title

Signal Processing Letters, IEEE

Publisher

ieee

ISSN

1070-9908

Type

jour

DOI

10.1109/LSP.2007.899332

Filename

4351938