Title :
IMM-based estimation for slowly evolving environments
Author_Institution :
Sch. of Electr. Eng., Seoul Nat. Univ., South Korea
fDate :
6/1/1998 12:00:00 AM
Abstract :
We propose a new approach to environmental parameter estimation for robust speech recognition in adverse conditions. The proposed method is based on the interacting multiple model (IMM) technique widely used in the area of multiple target tracking. Through a number of continuous digit recognition experiments, we can find the effectiveness of the IMM-based approach in slowly evolving environment conditions.
Keywords :
Gaussian distribution; Kalman filters; filtering theory; parameter estimation; speech recognition; IMM-based estimation; Kalman filtering; adverse conditions; continuous digit recognition experiments; environmental parameter estimation; interacting multiple model; robust speech recognition; slowly evolving environments; Gaussian distribution; Gaussian noise; Parameter estimation; Piecewise linear approximation; Robustness; Speech recognition; State-space methods; Target tracking; Taylor series; Working environment noise;
Journal_Title :
Signal Processing Letters, IEEE
