431336

Stochastic Integration and Long Term Predictor Estimation under Noisy Conditions for Speech Enhancement

Kuropatwinski, M. ; Kleijn, W.B.

Signal, Sensors & Syst. Dept, R. Inst. of Technol., Stockholm, Sweden

1

2005

March 18-23, 2005

801

804

Kalman filters; Wiener filters; least mean squares methods; maximum likelihood estimation; prediction theory; signal sampling; speech enhancement; stochastic processes; Kalman filter; LTP parameters; Wiener filter; a priori STP parameter distribution; clean speech estimates; databases; dual frame stationary process; excitation variances; lag; likelihood criterion; long-term predictor estimation; minimum mean square error estimates; noisy conditions; short term predictor; single frame stationary process; speech enhancement; speech training data sampling; stochastic integration; tap pairs; triple frame asymptotic mean stationary process; Covariance matrix; Databases; Kalman filters; Noise measurement; Sensor systems; Software measurement; Speech coding; Speech enhancement; Stochastic processes; Training data;

English

ieee

Acoustics, Speech, and Signal Processing, 2005. Proceedings. (ICASSP '05). IEEE International Conference on

1520-6149

0-7803-8874-7

conf

10.1109/ICASSP.2005.1415235

1415235