Journal article
Particle Filter Inference in an Articulatory-Based Speech Model
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.
| Language: | English |
|---|---|
| Publisher: | IEEE |
| Year: | 2007 |
| Pages: | 883-886 |
| ISSN: | 15582361 and 10709908 |
| Types: | Journal article |
| DOI: | 10.1109/LSP.2007.899332 |
| ORCIDs: | Winther, Ole |
Bandwidth Filtering Formant frequency Frequency Particle filters Reflection Signal processing Speech enhancement Speech processing Stability State estimation articulatory-based speech model auto-regressive coefficient autoregressive processes formant bandwidth formant frequency formant gain loose coupling parameter estimation particle filter inference particle filtering particle filtering (numerical methods) smooth time-varying parameter behavior speech enhancement time-varying auto-regressive speech model
