Markov-Switching GARCH Model and Application to Speech Enhancement in Subbands
- Author(s)
-
- Ari Abramson, (Department of Electrical Engineering, Technion)
- Israel Cohen, (Department of Electrical Engineering, Technion)
- Topics
-
- Sound enhancement and sound separation
- Noise reduction techniques
|
Get the paper in PDF format
Acrobat Reader
(version 5 minimum)
is necessary to read this document.
|
Abstract
In this paper, we introduce a Markov-switching generalized
autoregressive conditional heteroscedasticity (GARCH) model in the
short-time Fourier transform (STFT) domain. A GARCH model is utilized with
Markov switching regimes, where the parameters are assumed to be frequency
variant. The model parameters are evaluated in each frequency subband and a
special state (regime) is defined for the case where speech coefficients are
absent or bellow a threshold level. The problem of speech enhancement under
speech presence uncertainty is addressed and it is shown a soft voice
activity detector may be inherently incorporated within the algorithm.
Experimental results demonstrate the potential of our proposed model to
improve noise reduction while retaining weak components of the speech
signal.
|