Title :
Particle Filter Inference in an Articulatory-Based Speech Model
Author :
Beierholm, Thomas ; Winther, Ole
Author_Institution :
GN ReSound A/S, Ballerup
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;
Journal_Title :
Signal Processing Letters, IEEE
DOI :
10.1109/LSP.2007.899332
